Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
GP2GP
- Neil R Paul (23 May 2003)
-
Electronic record linkage and prevalence of diabetes
- Ritchie McAlpine, Alistair Emslie-Smith, Douglas Boyle, Andrew Morris for the DARTS/MEMO Collaboration (29 May 2003)
-
Re: Electronic record linkage and prevalence of diabetes
- Jeremy P Gray (29 May 2003)
-
Diabetes and the use of Read codes in the new quality and outcomes framework
- Clive L Morrison (30 May 2003)
-
Standards for Coding in General Practice
- Libby Morris, Robert Milne, Malcolm Campbell (2 June 2003)
|
|
|||
|
Neil R Paul, GP Partner The Commons Surgery, Sandbach, CW11 1HR
Send response to journal:
|
An important implication not mentioned by the authors is to do with the fact that many patients transfer between practices for a variety of reasons, including relocation and that this system is changing. Many of these patients will have diabetes, however the results are possibly valid for other diseases as well. In the last month a beta test of software has been announced as part of the GP2GP project, supported by most of the primary care computer suppliers. This will involve the electronic transfer of a patients record from practice to practice, a step to the electronic record. It will involve the transfer any existing codes to the new practice as well as information on consultations and other data. In theory this reduces the workload of each new practice which would orindarily have to resummarise the patients record when it eventually turned up. However the implications of this study are that the receiving practices shouldn't trust the data they are receiving because it may be inaccurate, imcomplete or coded differently to how the receiving practice would like it. Proceedures and protocols will have to be put in place to make sure that this is dealt with correctly and practices should be aware of this potential problem. Competing interests: None declared |
|||
|
|
|||
|
Ritchie McAlpine, Diabetes Regional Facilitator Ninewells Hospital, Dundee, DD1 9SY, Alistair Emslie-Smith, Douglas Boyle, Andrew Morris for the DARTS/MEMO Collaboration
Send response to journal:
|
Gray, et al report a diabetes prevalence of 2.53%, although the ascertainment date is not defined. The authors comment that a study using electronic record linkage in Tayside found a prevalence of diabetes of 1.9% - lower than theirs. We wish to point out that the prevalence figure in the Tayside study was for 1st January 1996. Our study showed that electronic record linkage of a variety of independent data sources gave a sensitivity and positive predictive value of 0.96 and 0.95 respectively. This was more sensitive than general practice registers (sensitivity 0.91/positive predictive value 0.98) in identifying diabetes (1). Over the 6 years since our study was published, the prevalence of diabetes in Tayside (population ~385000) has risen to 2.90%, an average annual increase of 0.16%. (1) Morris AD, Boyle DIR, MacAlpine R, Emslie-Smith A, Jung RT, Newton RW, MacDonald TM for the DARTS/MEMO Collaboration. The diabetes audit and research in Tayside, Scotland (DARTS) study: electronic record- linkage to create a diabetes register. BMJ 1997;315: 524-8. Competing interests: None declared |
|||
|
|
|||
|
Jeremy P Gray, Author of paper Battersea Research Group
Send response to journal:
|
We thank Dr McAlpine for his contribution. Our data was collected during 2001 which would make our prevalences very similar. Jeremy Gray Competing interests: None declared |
|||
|
|
|||
|
Clive L Morrison, General Practitioner Pendyffryn Medical Group, Prestatyn, Denbighshire. LL19 9DH
Send response to journal:
|
Gray et al (1) are presumably unaware of all the facilities and functions that are available on current General Practice clinical IT systems. Software such as Vision provided by In Practice Systems has a register for chronic diseases such as diabetes independent from Read codes. In my group practice, 695 patients are on the computer diabetes register with 364 (52.4%) of these having a C10 Read code for diabetes. The register negates the need to use the author’s incomplete and complex hierarchical Read code and medicine search criteria that would significantly underestimated our diabetic population. Software suppliers also provide within their systems embedded templates and guidelines on Diabetes Management Plans (2). These are in a structured format so that data is entered in a logical and comprehensive manner during a consultation, without the need for the clinician to make decisions on choosing the most appropriate of over 95 possible Read codes in the C10 hierarchy. In response to the new GMS contract negotiations, the software companies have already rewritten their programmes to take into account the quality and outcome frameworks for chronic disease management. The clinical IT systems will be updated once the contract has been agreed and if they are followed in practice will automatically standardise data entry, providing accurate information for audit and to calculate income. From a primary care point of view one of the major hindrances in collecting quality data is that secondary care is reluctant to move from paper based recording and correspondence. Such communications contain valuable information but in a non-searchable free text format. Expensive and labour intensive add on IT functions are available which can scan documents and then use OCR software to identify and extract data that can be recorded in Read codes. A more innovative project in Birmingham (3) has tried to address this problem differently by developing an electronic diabetes database and register, which can be accessed via an intranet system using an internet browser. This approach provides a single source of information and a shared-care view of data collected between the hospital and primary care of a common diabetes dataset. The system is live and data can be updated in real time. The database can be made available on a secured Internet site to allow patients to view reports and graphical representations of their management and results. References 1 Gray J, Orr D, Majeed A. Use of Read codes in diabetes management in a south London primary care group: implications for establishing disease registers. BMJ 2003; 326: 1130-2 2 In Practice Systems. Diabetic care. www.inps.co.uk/software/diabeticcare/_diabeticcare/fdiabeticcare.htm (accessed 30 May 2003) 3 Audit Commission; commissioning diabetes services. Birmingham Health Authority Diabetes Information Service. www.diabetes.audit- commission.gov.uk/CASESTUDIES/examples/birmingham4.htm (accessed 30 May 2003) Competing interests: None declared |
|||
|
|
|||
|
Libby Morris, General Practitioner Hermitage Terrace, Edinburgh, EH10 4RP, Robert Milne, Malcolm Campbell
Send response to journal:
|
Dear Sir, Re:
BMJ 2003;326:1130 (24 May)
We take issue with the conclusion of the authors that ' The use of Read codes for diabetes needs to be standardised and coding levels improved if valid diabetic registers are to be constructed and the quality of care is to be monitored effectively.' In fact, we have shown that standard read codes for recording disease already exist and are widely used in primary care .Many software systems used by GP practices to record clinical information already use standard data entry screens to help clinicians to manage patients with chronic diseases such as diabetes, asthma and coronary heart disease. Templates provide clinicians with reminders to check details such as blood pressure, cholesterol, HbA1c etc, and automatically record information using standard read codes, meaning that clinicians never need to see or have to choose a read code. Over 94% of GPs in Scotland now use a computer system to record clinical information, and 36% use it to record their management of chronic disease (1). In addition, clinical criteria for template screens have been developed by the Clinical Effectiveness programme of the RCGP in Scotland, (SPICE –pc) and are thus national standards. (2) A group representing users of all the clinical systems used in primary care in Scotland has been formed to agree on standard codes and templates for use in practices, called SCIMP (Scottish Clinical Information Management in primary care). A minimum set of codes for use in primary care was agreed in 1999 and are now in widespread use across Scotland, providing a national standard for the recording of diagnostic codes. (3) The SCIMP group recently carried out a study to investigate whether practices will be ready for the data reporting requirements for the new contract using CHD as an example, looking at electronic GP records of all CHD patients in five Scottish Practices validated by manual searches in 50 randomly selected patients in each practice. The study measured the recording of family history, smoking status, BP, diabetes testing, aspirin therapy and cholesterol measurement. The results showed that it is extremely easy for practices with completely electronic patient records to extract a disease register. Extraction of a complete data set takes several days if it involves checking through paper records, whereas setting up and running a search from electronic records is possible in less than two hours. If practices use the same clinical system and identical data entry templates, the data can be directly compared. We have shown that in Scotland, there is a high level of testing and recording of all the important information regarding patients with recorded CHD, irrespective of whether practices have fully electronic, paper based records, or a mixture of the two. If practices have fully electronic patient records, the information can be extracted easily, but unless there is a standard template, the information can only be viewed in isolation and is of little value for comparative purposes. When information is entered onto clinical records using standard templates, it can be extracted and amalgamated electronically. In Scotland, the information from SPICE-pc templates is collected and reported by PCCIU (Primary Care Clinical Informatics Unit) at Aberdeen University. (3) In Spring 2003, information was collected from 173 practices, with data from 1,052,770 patients. Of these, 39,771 were shown to have CHD (4.4%), 23,485 had diabetes (2.6%), 46,679 had active asthma (5.2%) and 73,357 were hypertensive and aged >35. (14.5%) This information was sent by the practices by email or disc, taking on average 10 mins by email, or 30 mins by disc. A similar function is provided by the PRIMIS unit in England, which gives advice on data recording and standard queries or searches, and extracts reports and amalgamates results across practices. (4) References: [1] Morris L. Dumville J, Campbell L M, Sullivan F A survey of computer use in primary care: general practitioners are no longer technophobic but other primary care staff need better access. Informatics in Primary Care (2003) 11: 5-11 [2] SPICE-pc: www.ceppc.org/spice/index.shtml [3] SCIMP:www.ceppc.org/scimp/index.shtml [4] PRIMIS:www.primis.nhs.uk Yours sincerely, Dr Libby Morris MBChB, Chairman of SCIMP Dr Robert M Milne MBChB, FRCGP, Senior lecturer, University of Aberdeen, Director of PCCIU Dr L. Malcolm Campbell, Senior lecturer, University of Glasgow Director of Quality Standards, RCGP Scotland Address for correspondence Dr. Libby Morris, RCGP (Scotland), 25 Queen Street, Edinburgh, Scotland Competing interests: None declared |
|||