Roy Harper

An exploration of knowledge and attitudes related to pre-pregnancy care in women with diabetes

Diabet. Med. 27, 1385–1391 (2010)

Implementing autonomy in a diabetes management system.

We have developed a speech-based telemedicine system which enables patients with hypertension and type 2 diabetes mellitus to send frequent, home-monitored health data via the telephone to the point of care. The decision support module in the system was tested using data from a cohort of 10 patients generated over a two-year period. Results from the tests indicate that the system is effective in providing personalized feedback to the patient and in generating alerts for the clinical user. The work suggests that this method of care delivery is practical, informative, and may improve the efficiency of chronic health-care delivery by reducing costs and improving patient-physician communication between hospital visits.

Pregnancy Planning and Diabetes: A Qualitative Exploration of Women's Attitudes Toward Preconception Care

INTRODUCTION Seeking preconception care is recognized as an important health behavior for women with preexisting diabetes. Yet many women with diabetes do not seek care or advice until after they are pregnant, and many enter pregnancy with suboptimal glycemic control. This study explored the attitudes about pregnancy and preconception care seeking in a group of nonpregnant women with type 1 diabetes mellitus. METHODS In-depth semistructured interviews were completed with 14 nonpregnant women with type 1 diabetes. RESULTS Analysis of the interview data revealed 4 main themes: 1) the emotional complexity of childbearing decisions, 2) preferences for information related to pregnancy, 3) the importance of being known by your health professional, and 4) frustrations with the medical model of care. DISCUSSION These findings raise questions about how preconception care should be provided to women with diabetes and highlight the pivotal importance of supportive, familiar relationships between health professionals and women with diabetes in the provision of individualized care and advice. By improving the quality of relationships and communication between health care providers and patients, we will be better able to provide care and advice that is perceived as relevant to the individual, whatever her stage of family planning.

Evaluation of a DVD for women with diabetes: impact on knowledge and attitudes to preconception care

Diabet. Med. 29, 950–956 (2012)

Appraisal of a conversational artefact and its utility in remote patient monitoring

The escalating rise in chronic illnesses, coupled with the shift in population demographics and increasing sedentary lifestyle has placed current health resources under considerable strain. In order to extend care to a wider audience, methods for managing these patients via telecommunications technology should be safe, usable and affordable. This paper reports on the development of a management solution for patients with diabetes and co-existing hypertension. The proposed system, known as DI@L-log, is an intelligent, automated remote monitoring system which involves patients proactively in the care of their condition by using spoken dialogue technology. We discuss the preliminary findings of a recent study involving 5 hypertensive diabetic patients from the Ulster community hospitals trust (UCHT) in Northern Ireland and its potential use in augmenting care for the increasing number of chronic disease patients.

Feature selection and classification model construction on type 2 diabetic patient’s data

Diabetes is a disorder of the metabolism where the amount of glucose in the blood is too high because the body cannot produce or properly use insulin. In order to achieve more effective diabetes clinic management, data mining techniques have been applied to a patient database. In an attempt to improve the efficiency of data mining algorithms, a feature selection technique ReliefF is used with the data, which can rank the important attributes affecting Type 2 diabetes control. After selecting suitable attributes, classification techniques are applied to the data to predict how well the patients are controlling their condition. Preliminary results have been confirmed by the clinician and this provides optimism that data mining can be used to generate prediction models.

Automated Phone Capture of Diabetes Patients Readings with Consultant Monitoring via the Web

This paper reports on a collaborative project between clinicians at the Ulster Hospital and researchers at the University of Ulster to produce a Web-based system for monitoring patients with Type 2 diabetes. The typical method of recording measurements of weight, blood sugar, and blood pressure allows for minimal intervention for the consultants as the paper-based records can only be reviewed at three monthly intervals. The new system allows the patients to use their own phone to input and record their measurements using speech, while the web-based view allows consultants to monitor these patients on a daily basis and to respond to alerts generated by the system. Patients could access their own readings in a tabular form and graphically via the Internet.

Exploring the needs, concerns and knowledge of women diagnosed with gestational diabetes: A qualitative study

OBJECTIVE to explore the concerns, needs and knowledge of women diagnosed with Gestational Diabetes Mellitus (GDM). DESIGN a qualitative study of women with GDM or a history of GDM. METHODS nineteen women who were both pregnant and recently diagnosed with GDM or post- natal with a recent history of GDM were recruited from outpatient diabetes care clinics. This qualitative study utilised focus groups. Participants were asked a series of open-ended questions to explore (1) current knowledge of GDM; (2) anxiety when diagnosed with GDM, and whether this changed overtime; (3) understanding and managing GDM and (4) the future impact of GDM. The data were analysed using a conventional content analysis approach. FINDINGS women experienced a steep learning curve when initially diagnosed and eventually became skilled at managing their disease effectively. The use of insulin was associated with fear and guilt. Diet advice was sometimes complex and not culturally appropriate. Women appeared not to be fully aware of the short or long-term consequences of a diagnosis of GDM. CONCLUSIONS midwives and other Health Care Professionals need to be cognisant of the impact of a diagnosis of GDM and give individual and culturally appropriate advice (especially with regards to diet). High quality, evidence based information resources need to be made available to this group of women. Future health risks and lifestyle changes need to be discussed at diagnosis to ensure women have the opportunity to improve their health.

A Supervised Learning Approach to Predicting Coronary Heart Disease Complications in Type 2 Diabetes Mellitus Patients

A supervised machine learning approach that incorporates genetic algorithms (GA) and weighted k-nearest neighbours (WkNN) was applied to classify type 2 diabetes mellitus (T2DM) patients according to the presence or absence of coronary heart disease (CHD) complications. The investigation was carried out by analyzing potential risk factors recorded at the Ulster Hospital in Northern Ireland. A GA initialization technique that integrates medical expert knowledge was compared with traditional data-driven GA initialization techniques. The results indicate that the incorporation of expert knowledge provides only a small improvement of CHD classification performance compared with models based on data-driven initialization techniques. This may be due to data incompleteness and noise or due to the beneficial effects of treatment, which masks the complication of CHD in the dataset. Further incorporation of expert knowledge at different levels of the GA need to be addressed to improve decision support in this domain

Evaluation of Outcome Prediction for a Clinical Diabetes Database

Diabetes is a metabolic disorder which can be greatly affected by lifestyle. The disease cannot be cured but can be controlled, which will minimize the complications such as heart disease, stroke and blindness. Clinicians routinely collect large amounts of information on diabetic patients as part of their day to day management for control of the disease. We investigate the potential for data mining in order to spot trends in the data and attempt to predict outcome. Feature selection has been used to improve the efficiency of the data mining algorithms and identify the contribution of different features to diabetes control status prediction. Decision trees can provide classification accuracy over 78%. However, while most bad control cases (90%) can be correctly classified, at least 50% of good control cases will be misclassified, which means that current feature selection and prediction models illustrate some potential but need additional refinement.