Special Issue Volume 21 Number 4 and Volume 22 Number 1

Application of information technology in clinical diabetes care

Part 1. Databases, algorithms and decision support

Part 2. Models and education

E.D. Lehmann, Guest Editor Academic Department of Radiology, The Royal Hospitals NHS Trust, St. Bartholomew's Hospital, London EC1A 7BE, UK and Department of Imaging, National Heart and Lung Institute (Imperial College of Science, Technology and Medicine), Royal Brompton Hospital, London, UK. Contact Form - Click Here for "About the Editor" A "Critical Appraisal" of these two Special Issues appeared in Diabetes, Nutrition and Medicine, click here for details

Medical Informatics: Volume 21 Number 4 and Volume 22 Number 1

Medical Informatics, 1996; 21(4): Part 1. Databases, algorithms and decision support

Application of information technology in clinical diabetes care

Part 1. Databases, algorithms and decision support

Editorial

1996 is the 75th Anniversary of the discovery of insulin. In 1922 Banting and colleagues' keynote paper was published [1]. A year later Banting et al. [2] wrote:

"It is not always easy to adjust so that there is sufficient insulin to nullify post-prandial hyperglycemia and yet insufficient insulin to produce dangerous lowering of the blood sugar".

Not much has changed in 75 years. This two part Special Issue focuses on advanced information technology (IT) initiatives to help adjust insulin doses, and control blood glucose (BG) levels, as well as assist in the provision of modern-day diabetes care.

1. Clinical background

Diabetes is a lifelong disease characterized by raised BG levels, due to under production or reduced action of the hormone insulin. It is associated with debilitating and life-threatening later life complications such as heart attacks, strokes, amputations, renal failure and blindness. In 1993 the Diabetes Control and Complications Trial (DCCT) demonstrated that maintaining tight BG control in patients with insulin-dependent diabetes can delay the onset and slow the progression of these later life complications [3]. However the chief adverse event associated with the intensive insulin therapy necessary to maintain such tight control was a three-fold increase in the incidence of severe hypoglycemia. Because of concern over forcing the BG level too low and causing hypoglycemia - maintaining tight glycaemic control in routine clinical practice remains not without its difficulties - as Banting observed way back in 1923 [2].

2. Computing background

Endeavors to apply IT routinely in diabetes care have been attempted for many years. Significant advances have been made by a number of key researchers in the field, some of whom are contributing descriptions of their latest work to these Special Issues. However, despite the novel work undertaken, the impact of computers in diabetes care thus far has been relatively limited. Figure 1 summarizes the spectrum of IT applications relevant to clinical diabetes care. The acceptance scale indicates how widely such applications have been adopted into routine clinical practice. For example, BG meters (either with or without electronic memories) are well accepted tools. By contrast at the other end of the spectrum computerized decision support is not at all widely used, at present. Attitudes however appear to be changing. The impact of the DCCT cannot be overemphasized. It is now apparent that existing standards of diabetes care are not sufficient. Tightening glycaemic control really makes a difference to patient wellbeing - yet the facilities are not available to provide the sort of intensive insulin therapy applied in the DCCT in routine clinical practice.

3. The Special Issues

These Special Issues are not intended as, and cannot constitute, a complete all-inclusive compendium on diabetes computing - but rather should be regarded as a selection of 'hot' topics as of late 1996 / early 1997. The papers included can be broadly classified under the following headings: (i) databases, (ii) algorithms, (iii) decision support, (iv) models, and (v) education; the first three of these topics being covered in this volume, with the other articles appearing in Part 2 of the Special Issue [4].

4. Database systems

Initiatives to reduce the morbidity associated with diabetes include the St. Vincent Declaration [5] which is in part looking to establish monitoring and control systems using IT for both quality assurance of diabetes health-care provision and for laboratory and technical procedures in diabetes diagnosis, treatment and self-management. In this issue Kopelman and Sanderson overview some of the specific requirements of database systems for diabetes care - before demonstrating how their own application, 'DIAMOND', addresses these. Particular attention is devoted to the manner in which databases can assist in the audit required for the St. Vincent Declaration, as well as to help ascertain clinical outcome measures. The following paper by Engelbrecht et al. overviews a particular database application - a diabetes smart card - highlighting some of the requirements specific to this form of patient-held database. A preliminary evaluation of the 'DIABCARD' in a hospital setting is also presented. Efficient data analysis and decision making both depend heavily on the frequency and quality of communication between patients and healthcare professionals. The trilogy of database-related papers is concluded with an overview by Gomez et al. of their 'DIABTel' system. This focuses on an alternative distributed database architecture, offering a telemedicine based approach for diabetes care.

5. Algorithmic dose-by-dose advisory systems

Algorithms by their very nature cannot cope with situations not explicitly stated. However they can cater for a large variety of situations - and if they can offer 90% coverage this could provide significant practical benefit to a considerable number of patients. Continuing the telemedicine theme Albisser et al. describe their 'HumaLink' system (formerly called 'TeleDoc') which builds on earlier algorithmic-based work from the lead author. Initial safety and efficacy 'beta-testing' experience with this system from 124 diabetic patients and 80 diabetic controls in two US centers is also presented. Moving to patient held devices, Holman et al. overview their patient-oriented insulin regimen optimizer ('POIRO') - a handheld decision support system - and describe short term experience using the device with six patients. The algorithmic dose-by-dose control trilogy is completed with a description by Williams of an alternative laptop computer-based system. Extended experience from the use of the 'Apple Juice' algorithms by a single patient over a six- year period are presented.

6. Visit-by-visit decision support prototypes

In any compilation of papers on diabetes-computing there needs to be space to describe new prototypes. Two such novel developments are overviewed here. Deutsch et al. describe their 'UTOPIA' system (UTilities for OPtimising Insulin Adjustment) which applies a data-driven decision-making paradigm. In contrast to model-based strategies, data-driven approaches focus on data summarization and interpretation, attempting to extract meaningful patterns and trends from noisy self-monitoring blood glucose (SMBG) data. The final paper in this issue by Hernando et al. describes their 'DIABNET' decision support prototype which integrates qualitative and quantitative reasoning for the analysis of SMBG data and therapy planning in gestational diabetes.

7. What to read in Part 2?

In Part 2 of this Special Issue [4] trilogies of papers on the application of models for therapy planning and the use of computers for diabetes education are presented, along with an Extended Conference Report of a European Symposium on diabetes computing held earlier this year in Graz, Austria.

References

1. BANTING, F.G., BEST, C.H., COLLIP, J.B., CAMPBELL, W.R., and FLETCHER, A. (1922) Pancreatic extracts in the treatment of diabetes mellitus: a preliminary report. Canadian Medical Association Journal, 12, 141-146.

2. BANTING, F.G., CAMPBELL, W.R. and FLETCHER, A.A. (1923) Further clinical experience with insulin in the treatment of diabetes mellitus. British Medical Journal, 1, 8-12.

3. THE DIABETES CONTROL AND COMPLICATIONS TRIAL RESEARCH GROUP (1993) The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New England Journal of Medicine, 329, 977-986.

4. LEHMANN, E.D. (1997) Ed. Application of information technology in clinical diabetes care. Part 2. Models and education. Medical Informatics, 22, (1), 1-120

5. WHO/IDF EUROPE (1990) Diabetes Care and Research in Europe. The Saint Vincent Declaration. Diabetic Medicine, 7, 360.

Medical Informatics, 1996; 21(4): Part 1. Databases, algorithms and decision support

Contents - Part One

Application of database systems in diabetes care. P. G. Kopelman and A. J. Sanderson

DIABCARD-an application of a portable medical record for persons with diabetes. R. Engelbrecht, C. Hildebrand, E. Brugués, A. De Leiva and R. Corcoy

Telemedicine for diabetes care: the DIABTel approach towards diabetes telecare. E. J. Gómez, F. Del Pozo and M. E. Hernando

Diabetes intervention in the information age. A. M. Albisser, R. I. Harris, S. Sakkal, I. D. Parson and S. C. En Chao

Randomized controlled pilot trial of a hand-held patient-orientated, insulin regimen optimizer. R. R. Holman, A. D. Smale, E. Pemberton, A. Riefflin and J. L. Nealon

Insulin algorithms in the self-management of insulin-dependent diabetes: the interactive ‘Apple Juice’ program. A. G. Williams

UTOPIA: a consultation system for visit-by-visit diabetes management. T. Deutsch, A. V. Roudsari, H. J. Leicester, T. Theodorou, E. R. Carson and P. H. Sonksen

DIABNET: a qualitative model-based advisory system for therapy planning in gestational diabetes. M. E. Hernando, E. J. Gómez, F. Del Pozo and R. Corcoy

• Return to Top of Page

Medical Informatics, 1997; 22(1): Part 2. Models and education

Application of information technology in clinical diabetes care

Part 2. Models and education

Editorial

This journal issue concludes a two-part Special Issue on the application of information technology (IT) in clinical diabetes care. Part 1 of the Special Issue [1] contained papers describing the application of database, smart card and telemedicine systems in diabetes care, and reported on the use of algorithms and other IT prototypes for computer-assisted clinical decision support. In this volume trilogies of papers are presented on the use of models for therapy planning, and the application of computers for diabetes education.

1. Models for therapy planning

The potential use of models (in their widest sense) for improving glycaemic control is discussed in the first paper by Worthington, who as an engineer with diabetes brings a unique perspective to this field. The variety of control system techniques adopted to try and improve glycaemic control is expanded in the following article by Naylor et al. which describes the development of computer-based models for predicting the blood glucose (BG) response of individual diabetic patients. This approach involves the use of robust control systems theory and approximate (nominal) models complemented by a representation of the uncertainty structure. The third paper in the modelling-therapy planning trilogy changes tack from glucose-insulin interaction to the gastrointestinal tract and food absorption from the gut. Although dietary intake is vital to the management of patients with diabetes, modelling this is often regarded as too complex or unrewarding for most researchers. While the experiments described, performed by Worthington on himself, are simple - they are of particular interest because they do not appear to have been done before.

2. Education

Despite the advanced technology which is being directed to the measurement and storage of self-monitoring BG (SMBG) data many patients, even in the 1990s, appear poorly equipped to alter their therapy on the basis of such data. As poor glycaemic control is associated with an increased later life risk of so many devastating complications it seems surprising that more effort has not gone into educating diabetic patients about what to do with their BG readings.

The paper, by Lehmann, addresses this point - overviewing the application of diabetes simulators, in particular those based on compartmental models, for use in the education of healthcare professionals, students, patients and their relatives. The purpose of these tools is to create a learning environment for communicating and training intuitive thinking when dealing with insulin dosage, dietary and lifestyle adjustments. The issue of evaluating educational interventions is addressed in the following paper by Brown et al. which overviews an interactive computer game ('Packy & Marlon') designed to improve self-care among children with diabetes. A randomized-controlled trial of 'Packy & Marlon' in 31 diabetic patients and 28 diabetic controls from two US centers is presented. The final paper in the education trilogy, by Day et al., overviews an alternative type of learning tool - a multi-media based educational package for patients, carers and healthcare professionals. One of the strengths of this system lies in its breadth of coverage, a whole host of practical information about diabetes and nutrition being provided in an easy to access interactive manner on CD-ROM.

3. Extended conference report

The final article in the Special Issue does not describe a single computer prototype or piece of research, but rather overviews a European Symposium on diabetes computing - held in the Autumn of 1996 in Graz, Austria [2]. Highlights from the meeting are presented, demonstrating some of the other novel work being undertaken in this rapidly expanding field.

4. Discussion

As should be apparent from the wide variety of papers included in these Special Issues, the diversity of approaches being tried, effectively to solve a single clinical problem - maintaining glycaemic control - is intriguing. However what conclusions can be drawn about the application of IT in clinical diabetes care, as of late 1996/early 1997?

BG meters and insulin pumps are well established tools. The use of database software is also widely accepted now. Furthermore, a range of educational games and programs for patient use at home are now available commercially. Some of these have started to be formally evaluated - but many have not. Decision support research has not yet come to fruition although promising approaches are starting to be tested. In this respect, most of the papers in the Special Issues do not propose, at present, to manage complicated cases using computer-based decision making tools - i.e without human input. Rather the consensus seems to focus on handling reasonably straight-forward clinical cases with a computer, thus freeing up more of health-care professionals' limited time to devote to the more difficult cases.

How can further advances be ensured? The way forward for decision-support and education in diabetes care is likely to be through integrated IT developments built on collaboration. For example, many of the individual components - required to build a state-of-the-art integrated system for clinical diabetes care - are already readily available now. All that appears to be lacking is a concerted effort to assemble - not just stand-alone prototypes - but open access, integrated systems. Particular attention however will need to be devoted to evaluation issues.

Although these Special Issues should not be considered in any way comprehensive in their coverage of clinical diabetes computing - it is hoped that the compilation of papers may provide a useful source of novel ideas - as well as, perhaps, a starting point for future research.

• Return to Top of Page

References

1. LEHMANN, E.D. (1996) Ed. Application of information technology in clinical diabetes care. Part 1. Databases, algorithms and decision support. Medical Informatics, 21, (4), 255-378

2. LEHMANN, E.D. (1996) Electronic medical records help diabetes care. Lancet, 348, 676.

Medical Informatics, 1997; 22(1): Part 2. Models and education

Contents

Controlling blood glucose: insights from an engineering control systems perspective. D. R. L. Worthington

Comparison of parameterized models for computer-based estimation of diabetic patient glucose response. J. S. Naylor, A. S. Hodel, A. M. Albisser, J. H. Evers, J. H. Strickland and D. A. Schumacher

Minimal model of food absorption in the gut. D. R. L. Worthington

Interactive educational simulators in diabetes care. E. D. Lehmann

Education video game for juvenile diabetes: results of a controlled trial. S. J. Brown, D. A. Lieberman, B. A. Gemeny, Y. D. Fan, C. M. Wilson and D. J. Pasta

‘Learning Diabetes’-a multi-media learning package for patients, carers and professionals to improve chronic disease management. J. L. Day, G. Rayman, L. Hall and P. Davies

Extended conference report: Computers in Diabetes ‘96. E. D. Lehmann

• Return to Top of Page

About the Editor

E.D. Lehmann, Guest Editor

Academic Department of Radiology, The Royal Hospitals NHS Trust, St. Bartholomew's Hospital, London EC1A 7BE and Department of Imaging, National Heart and Lung Institute (Imperial College of Science, Technology and Medicine), Royal Brompton Hospital, London, UK.

Dr. Lehmann received his medical training at Guy's and St. Thomas' Hospitals in London. He received his Bachelor of Science degree with Honors and Distinction in his Radiological Sciences research project in 1990, and was awarded his Bachelor of Medicine and Bachelor of Surgery degrees in 1993 - winning the Dean's Prize from the United Medical and Dental Schools (University of London) for his clinical research project. Dr. Lehmann has a longstanding interest in diabetes and computing. Prior to going to medical school he worked for a while for IBM (UK) Ltd in London and at their UK Headquarters in Portsmouth.

While studying medicine Dr. Lehmann returned to IBM to work part-time and during vacations on a European Community collaborative venture (the EURODIABETA project) researching the application of computers in diabetes care. Since then he has continued to develop his interests in medical informatics and diabetes computing. His work has included the development of an interactive simulator of glucose-insulin interaction in type 1 diabetes mellitus. This system, refined and updated, and linked to a knowledge-based system, database and dedicated data entry screen provides the AIDA simulator - an interactive educational diabetes program - which has been made widely available, without charge, as freeware software on the Internet. Since AIDA's first launch on the Internet in March 1996 over visits have been logged at the AIDA Web pages and over copies of the program have been downloaded (see http://www.2aida.org on the World Wide Web).

Dr. Lehmann has published extensively on the subject of computers in diabetes. In 1991 he was awarded the Peter Reichertz Memorial Prize for the Best Paper by a Young Scientist at MIE'91 (Medical Informatics Europe 1991), the Tenth International Congress on Medical Informatics held in Vienna, Austria - for a description of the clinical evaluation of a knowledge-based system for insulin therapy adjustment (see Medical Informatics 1993; 18(2): 83-101).

Since then Dr. Lehmann has been an invited speaker at many international meetings and has served as a referee for over half a dozen different medical journals. Dr. Lehmann recently authored a major 2 part systematic review of the literature on the application of computers in diabetes care for Medical Informatics (1995; 20(4): 281-302 and 303-329).

Dr. Lehmann currently is training in Diagnostic Radiology at The Royal Hospitals NHS Trust (St. Bartholomew's Hospital and The Royal London Hospital) in London. He is also enrolled for a part-time Ph.D. in the Department of Imaging at the National Heart and Lung Institute (Imperial College of Science, Technology and Medicine), Royal Brompton Hospital, London, UK where he is investigating biophysical properties of the arterial system in vivo.

Dr. Lehmann is always happy to talk to people about his interest in diabetes-computing. He can be contacted directly using the on-line AIDA contact form.

• Return to Top of Page

Figure 1, Medical Informatics 21 (4), 256: Spectrum of information technology applications in clinical diabetes care. The acceptance scale indicates how widely such applications are adopted into routine clinical practice. Education and decision support could be either for patients or healthcare students / professionals. (BG = blood glucose; KBS = knowledge based system CPN = causal probablistic network).

• Return to Top of Page