Insulin-like growth factor-I and insulin-like growth factor binding protein-1 in a representative population of type 2 diabetic patients in Sweden.

OBJECTIVE: To study the influence of type 2 diabetes on the insulin-like growth factor-I (IGF-I) and insulin-like growth factor binding protein-1 (IGFBP-1) serum levels in an area-based population of type 2 patients previously described. RESULTS: The patients (n = 151) were elderly (70.6 +/- 0.7 years of age) and moderately overweight (BMI 27.0 +/- 0.4 kg/m2). Most patients (83%) were treated with either diet alone or diet in combination with sulphonylurea. Metabolic control measured as HbAlc deteriorated with duration (p < 0.001) and between groups treated with diet (HbA1c 5.8 +/- 0.6), sulphonylurea (7.5 +/- 0.2) and insulin (7.7 +/- 0.4). Mean levels of IGF-I were within reported normal range, but were lower in the insulin-treated as compared to the non-insulin-treated patients. Levels of IGF-I decreased with diabetes duration and with increased blood glucose. There was a positive correlation between IGF-I and insulin levels and also an inverse correlation between IGF-I and IGFBP-1 levels. The IGFBP-1 levels were twofold higher than reported in non-diabetic individuals. In multiple stepwise correlation analysis, 37% of the variability in IGFBP-1 could be explained by BMI, IGF-I SD score, age, IGF-I, and fasting blood glucose. CONCLUSION: Our study indicates that influence of type 2 diabetes on IGF-I bioavailability in individual patients is modulated by insulin, body weight (presumably reflecting insulin sensitivity) and metabolic control. Furthermore, increased levels of IGFBP-1 are strongly associated with decreased b-cell function in type 2 diabetes mellitus.

A shared computer-based problem-oriented patient record for the primary care team.

1. INTRODUCTION. A computer-based patient record (CPR) system, Swedestar, has been developed for use in primary health care. The principal aim of the system is to support continuous quality improvement through improved information handling, improved decision-making, and improved procedures for quality assurance. The Swedestar system has evolved during a ten-year period beginning in 1984. 2. SYSTEM DESIGN. The design philosophy is based on the following key factors: a shared, problem-oriented patient record; structured data entry based on an extensive controlled vocabulary; advanced search and query functions, where the query language has the most important role; integrated decision support for drug prescribing and care protocols and guidelines; integrated procedures for quality assurance. 3. A SHARED PROBLEM-ORIENTED PATIENT RECORD. The core of the CPR system is the problem-oriented patient record. All problems of one patient, recorded by different members of the care team, are displayed on the problem list. Starting from this list, a problem follow-up can be made, one problem at a time or for several problems simultaneously. Thus, it is possible to get an integrated view, across provider categories, of those problems of one patient that belong together. This shared problem-oriented patient record provides an important basis for the primary care team work. 4. INTEGRATED DECISION SUPPORT. The decision support of the system includes a drug prescribing module and a care protocol module. The drug prescribing module is integrated with the patient records and includes an on-line check of the patient's medication list for potential interactions and data-driven reminders concerning major drug problems. Care protocols have been developed for the most common chronic diseases, such as asthma, diabetes, and hypertension. The patient records can be automatically checked according to the care protocols. 5. PRACTICAL EXPERIENCE. The Swedestar system has been implemented in a primary care area with 30,000 inhabitants. It is being used by all the primary care team members: 15 general practitioners, 25 district nurses, and 10 physiotherapists. Several years of practical experience of the CPR system shows that it has a positive impact on quality of care on four levels: 1) improved clinical follow-up of individual patients; 2) facilitated follow-up of aggregated data such as practice activity analysis, annual reports, and clinical indicators; 3) automated medical audit; and 4) concurrent audit. Within that primary care area, quality of care has improved substantially in several aspects due to the use of the CPR system [1].

Relationships between diabetes duration, metabolic control and beta-cell function in a representative population of type 2 diabetic patients in Sweden.

To clarify whether metabolic control and beta-cell function deteriorate with increasing duration of diabetes, we investigated in a cross-sectional study Type 2 diabetic patients in an area-based population. Type 2 diabetic patients (n = 231: 112 males, 119 females) were identified by age at onset > or = 35 years, fasting levels of C-peptide > 0.04 nmol l-1, and absence of islet cell antibodies. Body weight was slightly elevated (BMI 26.8 +/- 0.3 kg m-2), however 76/210 (36%), had normal weight (BMI < 25 kg m-2). Fasting blood glucose rose significantly during the first 10 years of known diabetes from 8.2 +/- 0.3 mmol l-1 in patients with 0-5 years of duration to 9.9 +/- 0.7 mmol l-1 in those with 5-10 years of duration, p < 0.01 and HbA1c from 6.4 +/- 0.2 to 7.4 +/- 0.4%, p < 0.05. Fasting C-peptide levels decreased after 10 years duration from 0.90 +/- 0.06 nmol l-1 during 5-10 to 0.69 +/- 0.08 nmol l-1 during 10-15 years of diabetes, p < 0.05. The proportion of insulin treated patients increased from 13% (12/94) with 0-5 years of duration to 33% (13/39) with 10-15 years and 60% (18/30) with more than 15 years of duration. In conclusion in Type 2 diabetic patients without signs of autoimmunity, metabolic control, and beta-cell function deteriorate with increasing duration of diabetes, leading to common but not inevitable occurrence of 'secondary failure'.

Experiences from the use of data-driven decision support in different environments.

The paper reports experiences from the implementation and use of data-driven decision support based on the Arden Syntax in three different environments: in a health-care centre, in a clinical laboratory, and in a research department. Methods and tools used for realization of decision support systems (DSSs) are shortly presented, and integration aspects of the DSS with running clinical applications and existing patient databases are discussed. The application areas are also described, together with validation procedures for the developed knowledge modules.

Drug interactions in primary health care. A retrospective database study and its implications for the design of a computerized decision support system.

OBJECTIVE: To investigate the occurrence of potential drug interactions in primary health care from the perspective of the prescribing general practitioner. DESIGN: Retrospective database study of computer-based patient records with a query language. All drug prescriptions during a four year period were compared with concurrent or overlapping prescriptions for the same patient and these drug pairs were compared with a database of drug interactions from the Swedish drug catalogue. SETTING: One health centre in Sweden with six general practitioners and two doctors on vocational training. PARTICIPANTS: All patients who had visited a doctor at the health centre between 1 November 1986 and 31 October 1990. MAIN OUTCOME MEASURES: The rate of potential interactions in relation to all drug prescriptions and the incidence rate of potential interactions for patients at risk (those receiving two or more drugs). RESULTS: Approximately 55,000 drug prescriptions were analysed for potential drug interactions. A total of 1,074 cases of potential drug interactions were found, which corresponds to a rate of 1.9% of all drug prescriptions. The incidence rate of potential interactions was 12% for all patients at risk (those receiving two or more drugs) and 22% for elderly (> = 65 years of age) patients at risk. Major interactions were investigated concerning the extent to which the prescribing doctors were aware of the potential interactions. CONCLUSION: Potential drug interactions occur at a high rate in general practice, in particular for elderly patients. Properly designed computer-based decision-support might increase the prescribing doctor's awareness of clinically significant interactions and improve the quality of drug treatment.

Decision support for drug prescription integrated with computer-based patient records in primary care.

A conceptual model of an information system that integrates a controlled vocabulary, a patient database, and a knowledge base is described. Methods, design and components for the implementation of the system are discussed. It is argued that the key issue for the successful introduction of computer-based decision support in primary care today is integration with a computer-based patient record. Also important is that the knowledge acquisition process is based on the general practitioner's real needs. This has been achieved by, first, providing general practitioners with real patient data from a series of retrospective database studies; and second, letting a panel of general practitioners select, discuss and decide which computer reminders to implement. A hybrid representation scheme was chosen for the knowledge base. The combination of a standard procedural representation (the so-called Arden syntax) for the reminder knowledge with a semantic net representation for the medical factual knowledge facilitates knowledge sharing with other systems and knowledge reuse within the system.

Medical audit based on computer-stored patient records exemplified with an audit of hypertension care.

OBJECTIVE: To audit hypertension care at a health centre using computer-based patient records as the source of information and a query language as the analysis tool. DESIGN: Retrospective database study comparing hypertension care in 1989 with hypertension care in 1990. SETTING: One health centre in Sweden with six general practitioners and two doctors on vocational training. PARTICIPANTS: All patients with hypertension in 1989 and 1990. MAIN OUTCOME MEASURE: The percentage of records that complied with the criteria in the hypertension care protocol. RESULTS: 585 records in 1989 and 574 records in 1990 were reviewed automatically by a series of 30 database queries. The computer time needed for the review was eight hours. The first audit showed deficiencies in the management of hypertension, in particular concerning patient history taking and risk factor analysis. The second audit, after the introduction of the hypertension care protocol, showed some minor improvements in the recording and also an increased rate of well treated hypertensive patients. CONCLUSION: Computer-based patient records may facilitate the review of medical records that is needed in medical audit. The audit demonstrates the gap between optimal care and clinical reality.

Extracting knowledge from a large primary health care database using a knowledge-based statistical approach.

Clinical databases from automated medical records represent a growing resource for deriving new medical knowledge. In this study a large primary health care database was explored with respect to the association between hypertension and diabetes. Data collection was made with a query language, and data analysis performed with an interactive knowledge-based statistical tool, MAXITAB, employing a multivariate tabular analysis technique. In the study population of 6660 patients the prevalence of diabetes was almost three times higher for hypertensive patients than for those with no hypertension. Conversely, the prevalence of hypertension was 2.6 times higher for diabetic patients than for those with no diabetes. The results support the assumption of a relationship between hypertension and diabetes, although the question of causality between the two diagnoses remains unsolved. Knowledge-based statistical tools of this kind may be feasible for exploring large clinical databases and may result in new medical hypotheses, worthy of further investigation.

The data dictionary--a controlled vocabulary for integrating clinical databases and medical knowledge bases.

The medical information systems of the future will probably include the entire medical record as well as a knowledge base, providing decision support for the physician during patient care. Data dictionaries will play an important role in integrating the medical knowledge bases with the clinical databases. This article presents an infological data model of such an integrated medical information system. Medical events, medical terms, and medical facts are the basic concepts that constitute the model. To allow the transfer of information and knowledge between systems, the data dictionary should be organized with regard to several common classification schemes of medical nomenclature.