| Project Objectives / Summary:
This project developed a clinical information model to assist GPs manage their use of pathology tests in relation to diabetesellitus.
The key areas of innovation were:
- A partnering approach that harnesses the key competencies of GPs, pathology and the Divisions and includes broad consultation with stakeholders;
- An information model that supports decision making in pathology through patient outcomes analysis;
- Management of pathology outcomes driven by general practitioners; and
- Cooperative use of information technology to facilitate clinical information management in pathology.
The project was structured as follows:
- Qpath was funded as a randomised controlled trial. All GPs from each of the three participating Divisions were offered the opportunity to be part of the project. Participants were selected randomly from those who registered.
- Historical patient data was retrieved from the database of Mayne Health Western Diagnostic Pathology.
- Database analysis was designed to establish patient population data sets linked to referring doctors.
- Classification of patients within these data sets was performed using artificial intelligence-based software (LabWizard) supplied by Pacific Knowledge Systems. Data was collected over the twelve-month period of the trial.
- Key performance indicator reports were generated monthly. These reports allowed GPs to compare themselves with the other GPs participating in the program, as well as a control group. All reports were confidential, and were accessible only by the GP and the Qpath project manager.
- Reports were available for access via a confidential web site. Hard copy reports were also available.
- Follow up lists were also sent regularly to participating doctors. These lists included names of patients who "failed"against the KPIs and were designed to be used as recall lists by the doctors, if required.
- At the end of the trial period, individual and group data was evaluated to determine the impact of the program on compliance with guidelines.
Project Collaboration:
General Practice Divisions of Western Australia partnered with Western Diagnostic Pathology and three Western Australian GP Divisions (two urban and one rural) in developing the outcomes-based model for managing pathology utilisation.
Project Outcomes:
- Less than 50% of patients at risk for diabetes are being tested annually to exclude diabetes but this appears to be improving over time.
- Approximately 70-80 % of known diabetics are monitored for glycaemic control at least annually by the HbA1c test. This percentage appears to be improving.
- Approximately 60-70% of known diabetics are monitored for the complications related to lipid metabolism at least annually by a cholesterol test or full lipid profile. This percentage appears to be holding steady over time.
- Less than 30% of urban-resident diabetics and 45% of rural diabetics are monitored at least annually for incipient diabetic nephropathy by a urine albumin test. This percentage is increasing slowly over time.
- Approximately 55% of participating GPs indicated that there was a poor understanding of the use of urine albumin testing as a monitoring test for incipient diabetic nephropathy. The increasing rate of testing appears to be related to increasing knowledge of its value.
- Constraints on GP consultations and limitations of existing patient recall systems make it difficult for GPs to address the issues of diabetic risk factors with patients identifiably "at risk".
- In general, practice recall systems are not in place at all or not sufficiently sophisticated to enable follow-up of patients identified as requiring additional consultations or pathology tests for the diagnosis or management of diabetes. Further, there is no effective mechanism for following up those who have not responded to recall.
- GPs have responded with enthusiasm to the type of self-monitoring process described by Qpath. Feedback has been particularly positive about the value of specific data related to their own individual clinical practice.
- There is a belief amongst GPs participating in Qpath that the magnitude of the task of better diagnosis and management of diabetes overwhelms current resources. Current infrastructure for general practice does not support the increased testing and patient recall and follow up required to effectively increase compliance with testing guidelines.
- The collaborative approach fostered by Qpath, bringing GPs, Divisions of General Practice, pathology providers and clinical pathology software vendors together has achieved a greater understanding of the roles of each. It has provided greater appreciation of the pressures currently existing within General Practice. It has also generated some ideas for how the separate groups could work together in the future for mutual benefit of all, especially the patient.
- The delivery of aggregated feedback data via Internet was not widely embraced by GPs. The uptake of IM / IT has not fostered a willingness to change work habits sufficient to encourage GPs to use the Internet for access to practice audit information.
Recommendations:
Clinical Audit
The confidential clinical audit and self-reflection model for pathology testing should be extended:
To allow participation of all GPs; and
To include other National Health Priority areas.
Stakeholders, including Government, should explore alternative funding and resourcing options to support participation in clinical audit by GPs and clinical pathologists.
Recall Infrastructure for Diabetes
Stakeholders should explore infrastructure support for diagnosis of diabetes in the General Practice community. This applies in particular to patient recall, testing and follow up systems. This could include:
A national clearing house for diabetes-related testing, and/or
Local clinical support teams
The Qpath Key Performance Indicators be used as a baseline set of KPIs for General Practice. Where required these KPIs should be updated, endorsed by national bodies and promulgated;
Stakeholders should explore resourcing issues for providing on-going feedback to General Practitioners regarding their compliance with these KPIs in diabetes.
Quality Use of Pathology
- Further research into effective means of improving the quality use of pathology should be supported;
- Stakeholders should consider alternative funding mechanisms for clinical pathology to support a "quality use of pathology" focus.
Medical Informatics
Training and research in medical informatics for General Practitioners and pathologists should be supported;
A working party to consider "the role of informatics in clinical pathology and General Practice" should be established.
Disseminating Project Information:
Publications:
Final Report sent to:
-
Fremantle Regional Division of General Practice
-
Canning Division of General Practice
-
Eastern Goldfields Division of General Practice
-
Brisbane Inner South Division of General Practice
-
Mayne Health Western Diagnostic Pathology
-
National Divisions Diabetes Program through Joan Burns Program Manager
-
MoU Diabetes Working Group through John Fisher ADGP Board Member
-
Julie Thompson ADGP Chair
Conferences:
National Divisions Forum, Sydney, November 2001
Lessons / Assessment:
Qpath has generated some important findings. Most notable are:
- The high level of interest in confidential clinical audit among GPs;
- Lack of understanding of urine micro-albumin testing;
- A lack of capacity of general practice to accommodate the implications of current guidelines for pathology testing in diabetes. In particular, the recall and follow-up systems are insufficient to deal with the volumes of individuals who need to be tested; and
- The lack of uptake of Internet technology for clinical audit purposes.
GP Acceptance of Clinical Audit
Despite the limitations of GPs capacity to implement large-scale change in their behaviour, Qpath nonetheless found an overwhelmingly positive response among doctors to the type of data they were receiving. There is clearly a lack of such aggregated clinical audit data available in general practice.
The general perception of the usefulness of projects like Qpath was evidenced by the high scores in both the entry and exit questionnaires relating to the likelihood of improvements being made to both diagnosis & management of diabetes. This perception seemed to be confirmed in the exit questionnaire when 95% said they found it interesting, 89% directly useful and 97% would participate in something similar again.
Qpath found little precedent for the diabetes KPIs and report formats, and this enthusiastic response suggests that further work is needed to extend, refine and implement audit programs in diabetes. Further, it is logical that the concept be extended to other areas, perhaps other national health priority areas such as cancer and heart disease.
One surprising finding was the lack of recognition of the "Defuse Diabetes" campaign, launched several months prior to the beginning of Qpath and the subject of several mailings to all Australian GPs. Only 7% of GPs at entry confirmed having these diagnosis and management guidelines, despite an extensive mail-out campaign by the Federal Government. This highlights the difficulty of delivering information by mail to GPs. It was interesting to note that at exit interview, this percentage had risen to 38%.
Overall, the project provides useful lessons for initiatives being mounted in other areas e.g. Chronic Disease Management, Information Management and Quality Use of Pathology.
|
