Chair: Rosemary Knight (Director, Evaluation Section, General Practice Branch, Dept of Health & Family Services)
Putting evidence into practice: development and content of a critical care pathway (CCP) for Fractured Neck of Femur (FNOF)
Authors: Holt D, March L,Chamberlain A, D,Cumming R, Cameron I, Finnegan T, Kurrle S, Skinner J.
Background: Critical Care pathways are treatment plans which act as a timetable for the delivery of healthcare. They provide a benchmark for the care and outcomes of most patients and are particularly suited to high volume conditions with a relatively predictable clinical course. Traditionally they have arisen from consensus.
Objective: To develop and implement an evidence based CCP for the treatment of FNOF in two Northern Sydney Area Health Service acute hospitals. A long term objective was to have the pathway replace the notes for these patients.
Methods: Core best practice elements for the treatment of FNOF were derived from a Cochrane style systematic review. After considerable consultation with those staff involved with FNOF patients, two pathways which were site specific but retained the core best practice elements were piloted.
Results: The development of the core best practice elements was the easier phase of this study. The development of a CCP which retained these elements but was individual enough to be 'owned' by the hospitals involved, required negotiation and motivation skills and a considerable amount of time. We will present some results on the barriers and aids to implementation and how compliance with best practice has been monitored.
Evidence Based Policy
Author: Andrew Holmes, Ministry of Health, New Zealand
Abstract: Policy makers are key potential clients of evidence on effectiveness in health care interventions. They have needs and perspectives which should be understood by those seek to provide such evidence.
The experience of the National Advisory Committee on Health and Disability (National Health Committee) is described with respect to its programme to promote the use of evidence based guidelines as a major part of the language and culture in the health sector. In particular, the experience in developing and implementing tools for assessing clinical priority have shown major gaps in evidence of effectiveness. Nevertheless, for coherent pragmatic policy to be developed, the gaps must be filled with lower levels of evidence and different types than customarily summarised in Cochrane Collaboration systematic reviews.
Implications as to how Cochrane Collaboration work might become more immediately useful and relevant to policy makers are discussed.
Developing an evidence-based guideline for preventing and managing diabetic foot problems
Authors: Jeannine Liddle, Margaret Williamson (Centre for Clinical Policy and Practice, NSW Department of Health, Australia)
Abstract: Objective of this presentation: To describe and discuss the difficulties in developing an evidence-based guideline for preventing and managing diabetic foot problems when a systematic review of the evidence showed few good quality studies.
Design: An Expert Panel agreed on the key questions to be addressed in the review. The Expert Panel consisted of specialists in diabetes care - endocrinologists, diabetes educators, podiatrists, consumers, general practitioners, epidemiologists and representatives from the NSW Department of Health.
The three key questions to be addressed in the review were:
Discussion points:
Future directions for clinical practice guidelines in Australian general practice: results of a national survey
Authors: Leena Gupta, Jeanette Ward, Robert Hayward
Background: Clinical practice guidelines are systematically designed statements to assist decisions about appropriate health care for specific circumstances. While there has been increasing international literature exploring clinicians' perceptions of guidelines and dissemination strategies, to date there have been no local studies of Australian general practitioners' (GPs) perceptions of guidelines and their attitudes towards different dissemination strategies.
Objectives: To determine GPs' views about clinical practice guidelines, including the impact of previously developed guidelines, factors influencing use, preferred dissemination strategies and access to information technology.
Method: A survey was conducted in 1995 of a national random sample of 373 Australian GPs using an 11-page self-administered questionnaire.
Results: 286 GPs (77%) returned completed questionnaires. While 92% considered that guidelines were good educational tools, GPs' recall of 9 different guidelines ranged from 52-94%. There were significant differences in GPs' ratings of the credibility of different organisations in guidelines development. The factor identified by 88% of GPs as very important in deciding whether to follow a guideline was whether it was based on scientific evidence such as systematic reviews. Strategies nominated as likely to increase the adoption of guidelines included academic detailing by a nurse and financial incentives. While 59% preferred guidelines in one official manual, 29% indicated a preference for dissemination of such evidence-based information in more innovative formats such as computer software. However, only 25% had access to a personal computer with a modem and 9% had Internet access.
Conclusion: Australian GPs appear positive about the overall purpose of clinical practice guidelines and an evidence-based approach, although the dissemination of specific guidelines has been patchy. These findings have implications for the dissemination of guidelines incorporating the results of Cochrane systematic reviews. Local intervention studies are needed to understand how the work of the Cochrane Collaboration can be communicated effectively to promote evidence-based clinical practice in Australia.
The Development of a Research Database to Enhance Autonomy and Evidence-Based Practice in Public Health
Authors: Dobbins M, Hayward S, Ciliska D, Brunton G, Hayward R, Underwood J
Background: Public Health workers are increasingly looking to research for guidance in making clinical decisions, advocating for programs in a competitive health care environment. However, finding and retrieving available information is often difficult and time-consuming.
Objectives: This paper describes the development, maintenance, and dissemination methods of a user-friendly, computerized reference database of public health research literature. Primary goals of the database are to facilitate the production of systematic overviews and support evidence-based practice and decision-making.
Methods: A series of systematic searches for published and unpublished literature yielded over 3000 relevant articles which have been entered into the database. The literature includes primarily intervention studies with some background articles. A system of keywords related to public health interventions was developed and applied to every entry. This keywording system allows users to conduct searches and retrieve studies according to clinical topic, research design and quality, study population, intervention strategy, and outcome category. Background articles can also be searched by research methodology, outcome measurement, public health practice and management, and health care policy.
Dissemination: The database has been disseminated through presentations in local, national and international venues, newsletters and online services. Various methods to evaluate online site use are being developed.
Results: 40-50 requests are received and responded to annually from public health workers, managers, health organizations, and other researchers The database provides the literature management system for ongoing production of effectiveness overviews (7 completed).
Conclusions: Three years of search, retrieval and keywording of public health studies has resulted in a user-friendly resource of effectiveness research accessible to decision-makers in practice and policy. The number of requests received demonstrates the potential of the database to facilitate evidence-based practice. Further development and evaluation of dissemination and coordination methods is required.
A common descriptive dataset for shared evaluation (systematic review) of interventions to implement clinical practice guidelines.
Authors: Finn Boerlum Kristensen and the CHANGING PROFESSIONAL PRACTICE EC BIOMED 2 Concerted Action Partners*), Danish Institute for Health Services Research and Development (DSI), Denmark.
Background: Research issues raised by clinical practice guideline implementation cut across national and cultural differences. Coordination of several intervention experiments and systematic review of their results call for a common theoretical framework and descriptive dataset.
Objectives: To develop a common descriptive dataset for contextual, intervention, and perceptual factors in interventions to change professional practice.
Methods: A Concerted Action project (CA) was started in March 1996 by 8 partners in 6 countries who plan and conduct guideline implementation experiments. Within 6 months the CA will revise its proposed scientific framework for evaluation of studies and produce a mapping of the Partners' projects on the framework, a common taxonomy of interventions, context etc., a description of key-factors influencing change, and a common dataset to describe context, intervention and perceptions in full.
Results: The Partners' studies to change practice in health care all meet specified inclusion criteria. However, they address different health care practices, settings, and controlled interventions. The CA's mapping of the individual studies and the resulting dataset for a prospective systematic review of their results and data will be presented for discussion.
* The CA comprises (Scientists in charge in parantheses) DSI (FB Kristensen), STAKES, Helsinki, Finland (M Mäkelä), Mario Negri Institute, Milan, Italy (A Liberati), Northern Centre of Health Care Research, Groningen, Netherlands (F Haaijer-Ruskamp), Health Services Research Unit, Oslo, Norway (S Flottorp), Health Services Research Unit, Aberdeen, UK (J Grimshaw), Centre for Health Economics, York, UK (N Freemantle), RAND/European Center for Policy analysis, Delft, Netherlands (JP Kahan).
Use of randomised controlled trials (RCTs) in drug subsidisation decisions
Authors: Suzanne Hill, David Henry, Andrew Mitchell, David Newby
Background: Under the requirements for listing of new drugs on the Australian Pharmaceutical Benefits Scheme, companies are required to submit applications supporting the cost-effectiveness of the product for its proposed indication. Guidelines for these submissions have been available since 1992. There has been a recent change to the Guidelines, so that companies are now encouraged to submit an economic analysis based on trial data, as well as one based on economic modelling.
Objectives: (1) To assess the usage of RCTs in economic analysis reviewed by the Pharmaceutical Benefits Advisory Committee (PBAC). (2) To assess the impact of use of RCTs on listing decisions by the PBAC.
Methods: A database was set up that records all submissions to the PBAC since 1992. The database is being analysed for: the number and type of submissions; the proportion using RCTs alone as the basis of the economic analysis; the proportion for which no RCTs are available and so require modelled analysis; those presenting both RCTs and modelled analyses. Differences in cost-effectiveness estimates between the 2 types of analysis have also been recorded where possible.
Results: At least 170 submissions have been reviewed by the PBAC since the introduction of the Guidelines in 1992. The proportion using RCTs is increasing, and the difference between the outcomes of analyses using RCTs and those using modelled economic analyses will be presented.
Implications: Use of RCTs for making policy and resource allocation decisions is likely to increase. This project describes some of the important differences between trial-based and modelled estimates of the cost-effectiveness of new drug therapies.
Incorporating evidence-based medicine into a consensus technique
Authors: R Gavin, C Pearson, D Roberton. University Department of Paediatrics, Women's and Children's Hospital, Adelaide.
Background: The Delphi Technique is a consensus method used to obtain the opinions of experts in a systematic way. The essential features of the Delphi Technique are: anonymity, feedback and repetition. Questions are formulated on the issues to be addressed and sent to a group of experts. The results are tabulated and reported to the respondants who then answer the questionnaire again. This process continues until opinion converges or 'fatigue' occurs. We report the effect of incorporating evidence from the literature into the Delphi Technique.
Method: We used the Delphi Technique to gain consensus about controversial management issues in acute gastroenteritis in children at the Women's and Children's Hospital in 1995. Respondants were provided with a critique of the relevant clinical trials from the medical literature at the same time as the results of the first round.
Results: The questionnaires were sent to 73 paediatricians and paediatric registrars with return rates of 82%, 82% and 77% for the three rounds. Consensus was reached on 34 of the 106 questions in the first round, and 34 of 72 questions in the second round. After three rounds four questions remained on which agreement could not be reached. Of the 68 questions which were repeated, unchanged, in the second round information was provided in the literature summary for 34. For questions for which no evidence was available, 41% showed a change in response favouring the view expressed by the majority of respondants in round one. Of the questions for which information was provided, 71% showed a change in response towards the views supported by the evidence. This difference is significant (p <0.05).
Conclusions: The Delphi Technique can be used as a tool for incorporating evidence-based medicine into practice. In this study evidence from the scientific literature influenced clinicians more than the opinions of their colleagues.
This page was last updated on 18 November 1996
Comments for improvement or correction are welcome.
georg.koch@cochrane.de