Development and validation of a risk-score model for subjects with impaired glucose tolerance for the assessment of the risk of type 2 diabetes mellitus—The STOP-NIDDM risk-score
Introduction
It is now widely appreciated that we are experiencing a global epidemic of type 2 diabetes. In 2003, the International Diabetes Federation (IDF) estimated that 194 million people worldwide had diabetes; this figure is expected to increase to more than 300 million by 2025 [1]. The prevalence of prediabetes is estimated to be even higher, and increasing at a similar rate [1]. Beating this epidemic will require widespread action to increase the early identification and treatment of dysglycaemia, especially in people likely to progress to type 2 diabetes [2]. Straightforward, inexpensive methods of identifying individuals at high risk of developing type 2 diabetes – individuals who would benefit most from treatment – are therefore extremely valuable [3] (Fig. 1).
In many other therapeutic areas of chronic diseases, the use of risk-prediction tools is common. There are, for example, numerous tools available to assess an individual's risk of developing cardiovascular disease (CVD), including the European SCORE [4] and PROCAM score [5], the New Zealand score [6], and the US Framingham 1991 and 1998 scores [7], [8]. Recently, a risk-score has also been developed to predict dementia on the basis of simple risk factors [9].
Such tools estimate an individual's risk of having a defined ‘event of interest’. In prediabetes, the event of interest is the transition from impaired glucose tolerance (IGT) to type 2 diabetes [10], [11]. There is also strong evidence to indicate that this progression can be halted or reduced by lifestyle or pharmacological interventions [12], [13]. Stern et al. have noted that the standard method of identifying individuals at high risk for type 2 diabetes is a somewhat costly and inconvenient 2-h oral glucose tolerance test (OGTT) to determine whether they have IGT [14]. In most cases, minimal or no attention is paid to the effects of other known risk factors for type 2 diabetes, such as advanced age and obesity.
Consequently, risk-scores have been developed for the prediction of type 2 diabetes that do take account of such factors, including the FINDRISC score [15], the American Diabetes Association diabetes risk test [16], the Thai risk-score [17] and several other scores derived from various European [18], [19], [20], [21], [22], [23], [24], American [25], [26], [27], [28], Indian [29] and Arab [30] population studies. These tools are based on traditional risk factors for type 2 diabetes. The evidence increasingly indicates that the effect of metabolic factors such as hyperlipidaemia and hypertension should also be considered [27], [31]. Such tools would allow wider screening, reduce the number of individuals with undiagnosed prediabetes or type 2 diabetes who remain undetected, and help to improve healthcare provision. It is well-known that people with IGT have a high probability to progress to diabetes; approximately half of them are diabetic within 10 years [32], [33]. The rate of progression among people with IGT depends on their risk factor profile, and thus type 2 diabetes risk-scores may provide a way to predict the risk also in people with IGT. This way it would be possible to identify those people with IGT who are at the highest risk and who may most benefit from preventive interventions, as demonstrated with the FINDRISC in the Finish Diabetes Prevention Study recently [34].
The objective of the present study was to develop and validate a simple model to predict the risk of development of type 2 diabetes in individuals with IGT.
Section snippets
Subjects
Data from the STOP-NIDDM study population were used to develop the model. STOP-NIDDM was a double-blind, placebo-controlled, randomised trial conducted in nine countries (Canada, Germany, Austria, Norway, Denmark, Sweden, Finland, Israel and Spain). Details of the study and its findings have been described elsewhere [35], [36], [37] and are briefly summarised below.
Participants were recruited from a high-risk population, aged between 40 and 70 years, with a body mass index (BMI) between 25 and
Results
A risk-score model was developed using a Cox PH approach based on data from 1160 individuals who completed STOP-NIDDM. The eight variables included in the final model were: the presence or absence of acarbose treatment, gender, serum triglyceride level, waist circumference, FPG level, height, baseline history of CVD, and diagnosed hypertension. Each of these variables was significantly associated with an individual's risk of type 2 diabetes (P ≤ 0.05). Descriptive statistics, regression
Discussion
The present manuscript reports the development of a simple and validated tool to estimate an individual's risk of developing diabetes in the next 2.5 years. The tool can be used to identify individuals with IGT at the highest risk of developing type 2 diabetes, who are likely benefit most from prevention strategies.
Importantly, the tool is based on commonly available demographic and clinical data. The eight variables included – acarbose treatment, gender, serum triglyceride level, waist
Conflict of interest
JT has received reimbursements of travel expenses, research grants and lecture fees from Bayer HealthCare. Martin Hellmich and Walter Lehmacher have been working on this project paid by Bayer; Torsten Westermeier, Thomas Evers and Andreas Brückner are empoyed by Bayer. The STOP-NIDDM trial where Jean-Louis Chiasson is the Principal Invsestigator was sponsored by Bayer. This present work was carried out by an educational grant from Bayer to the University of Helsinki.
Acknowledgements
This work was partially funded by an unrestricted grant from Bayer HealthCare. Editorial assistance was provided by Gardiner-Caldwell London.
References (46)
- et al.
Cardiovascular disease risk profiles
Am. Heart J.
(1991) - et al.
Risk score for the prediction of dementia risk in 20 years among middle aged people: a longitudinal, population-based study
Lancet Neurol.
(2006) - et al.
Diabetes risk score in Oman: a tool to identify prevalent type 2 diabetes among Arabs of the Middle East
Diabetes Res. Clin. Pract.
(2007) - et al.
Sustained reduction in the incidence of type 2 diabetes by lifestyle intervention: follow-up of the Finnish Diabetes Prevention Study
Lancet
(2006) - et al.
Acarbose for prevention of type 2 diabetes mellitus: the STOP-NIDDM randomised trial
Lancet
(2002) - International Diabetes Federation, Diabetes Atlas, Available at: http://www.eatlas.idf.org (accessed April...
The burden of type 2 diabetes: are we doing enough?
Diabetes Metab.
(2003)- et al.
Prevention of type 2 diabetes: issues and strategies for identifying persons for interventions
Diabetes Technol. Ther.
(2004) - et al.
Estimation of ten-year risk of fatal cardiovascular disease in Europe: the SCORE project
Eur. Heart J.
(2003) - et al.
Simple scoring scheme for calculating the risk of acute coronary events based on the 10-year follow-up of the prospective cardiovascular Munster (PROCAM) study
Circulation
(2002)
Updated New Zealand cardiovascular disease risk-benefit prediction guide
Br. Med. J.
Prediction of coronary heart disease using risk factor categories
Circulation
Relation of impaired fasting and postload glucose with incident type 2 diabetes in a Dutch population: the Hoorn Study
JAMA
Progression to clinically diagnosed and treated diabetes from impaired glucose tolerance and impaired fasting glycaemia
Diabet. Med.
The major diabetes prevention trials
Curr. Diab. Rep.
Prevention of type 2 diabetes mellitus
Identification of persons at high risk for type 2 diabetes mellitus: do we need the oral glucose tolerance test?
Ann. Intern. Med.
The diabetes risk score: a practical tool to predict type 2 diabetes risk
Diabetes Care
Diabetes Risk Test
A risk score for predicting incident diabetes in the Thai population
Diabetes Care
Performance of an NIDDM screening questionnaire based on symptoms and risk factors
Diabetes Care
Diabetes risk score: towards earlier detection of type 2 diabetes in general practice
Diabetes Metab. Res. Rev.
A Danish diabetes risk score for targeted screening: the Inter99 study
Diabetes Care
Cited by (32)
Development and validation of risk prediction models for new-onset type 2 diabetes in adults with impaired fasting glucose
2023, Diabetes Research and Clinical PracticeDetermining type 2 diabetes risk factors for the adults: A community based study from Turkey
2018, Primary Care DiabetesCitation Excerpt :Screening questionnaires are becoming more widely used because they include the main risk factors and can be easily conducted by trained personnel [4]. It is suggested that use of risk determination tools in social studies is useful both for determining persons facing risks and raising awareness about the modifiable risk factors [10–15]. Suggestions of the international health authorities indicate that all adults should be evaluated for T2D risks based on their demographic and clinical characteristics [16,17].
Relationship between the Finnish Diabetes Risk Score, fasting blood glucose and glycohemoglobin A1c
2023, Revista de la Sociedad Argentina de DiabetesDigital solution for detection of undiagnosed diabetes using machine learning-based retinal image analysis
2022, BMJ Open Diabetes Research and CareLower miR-21/ROS/HNE levels associate with lower glycemia after habit-intervention: DIAPASON study 1-year later
2022, Cardiovascular Diabetology