BEACH | Previous research
Bettering the Evaluation and Care of Health (BEACH)
The BEACH© program continuously collects
information about the
 |
patients seen |
|
|
reasons people seek medical care |
|
|
problems managed |
|
|
treatments provided in general practice in Australia. |
It uses a cross-sectional, paper based data collection system developed over the past twenty years in the Department of General Practice, University of Sydney. Data generated is used by researchers, government and industry.
See the BEACH page for more details and to download the latest reports.
Classifying ICPC-2 PLUS terms to ICD 10 and to ICD 10 AM.
ICPC-2 PLUS is currently being used by about 1500 GPs in about 400 practices in their electronic health records, and by a range of researchers for the primary or secondary coding and classification of morbidity managed in primary care.
When there is a need to compare data collected in ICPC-2 with other data from hospitals or mortality statistics (in ICD 10 AM, or in ICD 10), the data need to be in the same classification.
When the ICPC-2 PLUS terms are classified according to ICD 10AM,
users will be able to translate their PLUS terms into ICD10 AM as an alternative to the Australian
national primary care standard classification of ICPC-2.
Development of BEACH Statistical Evaluation Areas (SEAs)
Introduction: In the national BEACH program a sample of 1,000 GPs participate per year. The national data are widely published as the reference point about general practice activity.
However, GP Divisions need more localised data to assist them in evaluating the impact of their programs. We need a minimum of 40 participating GPs (on average) to describe the activities of a group of GPs (because of wide variance between individual GPs).
Few Divisions are of sufficient size to gain a minimum sample of 40 participating GPs per year. Exceptions are large Divisions such as Brisbane North.
Aim:
To develop BEACH Statistical Evaluation Areas (SEAs) that are made up of multiple Divisions that are similar in terms of their classification in:
in a manner that provides sufficient statistical power to make valid comparisons and monitor trends in the local area.
Method: Queensland was selected as the trial State. Each postcode was allocated fully/ partially to a Division. For each Division we considered:
Results: We experienced difficulties with postcode splits across Divisions, and wide variation in SEIFA, RRMA and ASGC classes within any single Division. We developed a draft proposal of five SEAs for Queensland. Queensland Divisions of General Practice (QDGP) considered the proposal, and based on local knowledge suggested some changes to the SEA groupings. These were adopted. The revised proposal was distributed to all Queensland Divisions (through the QDGP) with a request for criticisms and comments. The suggestions were considered, but none were adopted because of their negative effect on other SEAs.
The final SEAs for Queensland are summarised below. The other states and territories will be reviewed as time and funding permits.
| Rural | |
| 416 | Northern & Western Queensland Primary Health Care |
| 417 | Far North Queensland Rural Division of General Practice Assn Inc |
| 410 | Central Queensland Rural Division of General Practice Assn Inc |
| 412 | Townsville Division of General Practice Ltd |
| 413 | The Cairns Division of General Practice Ltd |
| 414 | Southern Queensland Rural Division of General Practice Assn Inc |
| Provincial | |
| 411 | Mackay Division of General Practice Ltd |
| 418 | Sunshine Coast Division of General Practice Assn Ltd |
| 419 | Capricornia Division of General Practice Ltd |
| 420 | Wide Bay Division of General Practice |
| Satellites | |
| 406 | Gold Coast Division of General Practice Ltd |
| 407 | The Redcliffe Bribie Caboolture Division of General Practice Assn |
| 408 | Ipswich and West Moreton Division of General Practice |
| 409 | Toowoomba and District Division of General Practice Ltd |
| Southern Metro | |
| 401 | The Assn of the Brisbane Inner South Division of General Practice In |
| 402 | The Brisbane Southside Central Division of General Practice Assn Inc |
| 403 | Association of Bayside GP Division (Brisbane) Inc. |
| 404 | Logan Area Division of General Practice Ltd |
| Brisbane North | |
| 405 | Brisbane North Division of General Practice Assn Inc |
OTHER RECENT AND CURRENT PROJECTS
Title: Mapping a
sample of ICPC-2 Plus terms to SNOMED-CT and gap analysis of the result
Year: 2005
Investigators: Britt H, Miller G, and O'Halloran J.
Funding source: National eHealth Transitional Authority
A gap analysis of terms and concepts contained within existing Australian terminologies but not contained in SNOMED-CT (SCT) was requested by the National E-Health Transition Authority to inform the national strategy of acquiring a licence for SCT.
Target terminologies were ICPC-2 Plus, Docle, ICD10AM (plus ICD10AM index and terms), CATCH, Aged care, Austin RMC, CIP and the University of Adelaide GP terminology. Both computer matching and manual mapping techniques were used. Computerised matching was done by a modified University of Adelaide method.
Computerised matching showed very poor congruence between SCT and most terminologies (3%-67%). However manual mapping using expert coders provided matching rates at level 1 or 2 of 45% to 91% (the latter for ICPC-2 Plus). These results included mapping using post coordination of SCT terms (10% for ICPC-2 Plus).
Computerised matching while theoretically tempting is a clearly ineffective way of mapping terms to SNOMED-CT. The success rate of mapping using coders underlines the importance of human interpretation in selection of concept matches between terminologies. This is particularly important in the interpretation of the vernacular terms sometimes used by clinicians.
The good congruence between SNOMED CT and GP terminologies in
common use in Australia means that adapting SNOMED for use in GP systems would be relatively
straightforward.
Title:
General Practice EHR and data query minimum data set development
Year: 2005
Investigators: Miller G, Britt H, and O'Halloran, J.
Funding source: General Practice Computer Group (Phase 3 Development Grants)
This project was commissioned by the Electronic Communication Working Group of the General Practice Computer Group and conducted under the auspice of the RACGP. This project developed a minimum set of data items necessary for reporting from GP computer systems. The data items were derived from established reporting data sets used in general practice in Australia. While these data items were derived from reporting sets, all the data items have relevance to the clinical activities of general practitioners.
The AGPSCC contacted custodians of data sets and negotiated supply of the lists and specifications of the data elements. These were in a variety of formats and data specification structures. The data sets were amalgamated to remove duplicate items and duplicates rationalised where differences existed between similar items. This created a preliminary minimum data set which was then circulated to the data custodians for comment. General agreement was reached on the content. This set was also circulated to the AIHW National Data Development and Standards Unit and the NeHTA Clinical Data Standards Group for comment. The Clinical Data Standards Group were also consulted regarding data elements and specification which they had developed which were similar to elements in the preliminary minimum data set. Supply of these data elements to this project was negotiated with NeHTA. It was decided to format the minimum data set in NeHTA format to facilitate use in other related projects. Research was undertaken to elicit standardised data definitions based on commonly used definitions relevant in the context of general practice.
The final minimum data set comprised 90 data elements and included data groups of logically associated items and a linkage diagram to specify required linkages between data items.
Further work is urgently needed to resolve the ‘missing’ value
domains in some of the data definitions in the minimum data set. This is closely related to
Government decisions regarding the purchase or development of a standard health terminology.
Discussion is needed between the various stakeholders, for example, GPCG, NeHTA, the HL7 group,
openEHR, AIHW and AGPSCC, to plan to resolve these problems. This problem must be solved before we
can proceed to final openEHR archetype and HL7 query implementation.
Title: Study of GPs working in Community Health Service in Victoria
Year: 2005
Investigators: Britt H, Miller GC
Funding Source: Victoria Department of Human Services
The Victoria Community Health Services (CHS) has become a major platform for the delivery of health services in Victoria. The role of GPs in Victorian CHS is inadequately defined. Systems for data collection are currently underdeveloped so there is a lack of reliable evidence to illustrate the impact of GPs in CHS.
This research aims to describe GP activity and profiles of patients attending Victorian CHS. Evidence based general practice data will help both Government and CHS management understand the unique clinical role and the characteristics of patients to whom CHS GPs provide their services.
This study utilises the methods of the BEACH program (Bettering
the Evaluation and Care of Health), a continuous data collection program now in its eighth year,
investigating general practice activity on a national level. BEACH has become recognised as the
definitive source of general practice data in Australia.
Title: Study of the clinical experience gained by Registrars
Year: 2005
Investigators: Spike N (Monash University), Britt H, Miller GC
Funding Source: Victorian Metropolitan Alliance
In collaboration with Monash University and the Victorian Metropolitan Alliance (VMA) this study aims to:
This study utilises the methods of the BEACH program (Bettering the Evaluation and Care of Health), a continuous data collection program now in its eighth year, investigating general practice activity on a national level. BEACH has become recognised as the definitive source of general practice data in Australia.
Current status: in the
field
Title: A comparison of primary care practice in the US, UK, New Zealand and Australia
Funding Year: 2004-05
Investigators: Bindman, A. B. (UCLA) Majeed, A. (Imperial College, London) Forrest, C. (Johns Hopkins) Crampton, P. (University of Otago) and Britt, H. (University of Sydney).
Funding source: Commonwealth Fund (US)
The original objectives of this study were to compare aspects of primary care practice in Australia, New Zealand, the United States, England and Canada, in order to provide insights into how different health care systems affect the clinical activities of primary care physicians.
In reality what we could compare was: annual average exposure of the population to primary care physicians; patient mix, scope of practice, visit duration, and specialty referral rates.
The comparison could only be made in 2001 because New Zealand only had a single year of data available for comparison.
We were limited to comparisons between Australia, New Zealand and the United States. The data from the UK General Practice Research Network proved unreliable for measurement of morbidity managed at the encounter because of a lack of linkage in the records between the encounter and the problems under management.
The data sources used in the comparison were:
Current status: A number
of papers are in preparation.
Title: FRACGP - does it make a difference?
Year: 2002
Investigators: Britt, H. and Miller, G. C.
Funding Source: Royal Australian College of General Practitioners
This study was a secondary analysis of BEACH data. Certification examination results have significant implications for career prospects and remuneration of candidates. Whether or not certification affects the care provided is rarely investigated. This study compares practice patterns of general practitioners in Australia who are vocationally certified and those who are not. We compared the characteristics of certified and uncertified GPs, their patients, encounters, problems managed and management actions, and tested 34 performance indicators. We used logistic and multiple regression to investigate whether differences identified were explained by other factors.
Findings: Of 1,975 GPs 659 (33.4%) held FRACGP. FRACGPs were more likely to be: female; younger; Australian graduates, working fewer sessions; in larger practices; in accredited practices; using computers for clinical purposes. Their patients were younger, more often female, less likely to hold a health care concession card. They claimed more long consultations and had longer timed consultations, prescribed fewer medications, provided more clinical treatments and procedures, ordered more pathology tests and referred more to other health professionals. After adjustment for GP/practice, patient and morbidity differences, FRACGPs still claimed more longer consultations, and after further adjustment for they still had longer measured consultation length, did more therapeutic procedures, prescribed less overall, prescribed fewer non steroidal anti-inflammatory drugs in the elderly and fewer antibiotics for upper respiratory infections.
Conclusion: Certification of general practitioners predicts some aspects of clinical care.
| Status: compete. | |
| G. Miller, H. Britt, Y. Pan, and S. Knox. Relationship between general practitioner certification and characteristics of care. Med Care 42 (8):770-778, 2004. Link to Publication. | |
| Note: See this and other publications in the Refereed articles in recognised journals list in the FMRC publications. | |
Title: A general practice computerised active data collection
validation study
Year: 2001-02
Investigators: Britt, H. and Miller, G. C.
Funding agency: Royal Australian College of General Practitioners and Western Sydney Division of General Practice
This study compared two means of data collection for BEACH: the paper-based method and an active computerised data collection method. Twenty-eight GPs who had done BEACH on paper took part in the study allowing direct comparison of the results of the two methods, without the influence of differences in the characteristics of two GP samples. This is the first time a study tested the extent to which this change in data collection method affected the data recorded. The aim was to assess whether computerised collection was acceptable to GP participants, whether it produced the same overall results as the paper based method, and the feasibility of this method of collection for future BEACH participants.
The results indicated that most of the GPs found active computer data collection more difficult than on paper, and there was a large decrease in the amount of information provided on computer when matched to the GP’s paper collection. It was concluded that at this stage active computerised data collection on a program running separate from the clinical software was not feasible as a data collection method for representative national general practice data (paper in preparation).
* Several other projects have started but have not been added to this page as yet.
Related Links
SNOMED CT
ICPC-2 PLUS
CAPS
ATC
Research Networks & Units
(Faculty of Medicine, The University of Sydney)
Research centres and institutes
(The University of Sydney)
Medicine
resources - directory of Medicine related websites.
Tasmanian GP Atlas - The GP Atlas
seeks to demystify the General Practice sector in Tasmania by providing a catalogue of relevant
on-line and other information resources for rural GPs, their family members, Practice Managers,
Practice Nurses, GP Registrars, prevocational doctors, medical and other students.
Primary Care Electronic
Library (PCEL) - Community Health Sciences, St. George's, University of London.
