The ADDITION Study. Intensive Treatment in People With Screen Detected Diabetes in Primary Care.

Aims:

• To develop and evaluate strategies for early detection of type 2 diabetes in different countries and different populations.
• To study whether a multifactorial treatment strategy can reduce CVD-mortality and reduce the incidence of macro- and microvascular complications. The treatment strategy consists of motivational interviewing, encouraging behavioural changes (dietary advises, physical activity, smoking cessation) and intensive pharmacological treatment of blood pressure, blood glucose, and serum lipids
• To identify genetic markers predicting development of diabetic complications
• To evaluate health economical consequences of screening and early intervention for type 2 diabetes

Design and methods:

The study is an investigator initiated and designed study, initiated in Denmark by the two principal investigators, planned and conducted in collaboration between the four centers in Denmark, UK and the Netherlands. The study has two elements: a screening study and a subsequent intervention study.

In the screening study, the feasibility and results of country specific models to identify undiagnosed individuals with Type 2 diabetes will be evaluated.

In the treatment study the effects of routine care in general practice according to local and national guidelines will be compared with an intensive ADDITION protocol, including structured lifestyle education (dietary modification, increased physical activity and smoking cessation) and intensive treatment of blood glucose, blood pressure and lipids, and prophylactic aspirin with or without motivational interviewing, on mortality, macrovascular and microvascular disease. Furthermore the impact of treatment on health status, treatment satisfaction and health service costs will also be assessed.

Methodology - Screening study:

In Denmark > 300 primary care physicians from 5 different counties (Copenhagen, Aarhus, Ringkøbing, Ribe and Sønderjylland) participate in the study. Diabetes-related information is sent to all individuals aged 40-69 years enrolled in their practice. A questionnaire (diabetes risk score29) including age, gender, family history of type 2 diabetes; obesity; physical activity and previously diagnosed hypertension was used. Individuals scoring high on the questionnaire are encouraged to contact their physician for an examination of random blood glucose (RGB) and HbA1c. A step-wise strategy based on RBG, HbA1c, FPG and OGTT is used to diagnose diabetes In the Netherlands all people aged 50-69 years and listed with the participating primary care physicians is invited to fill in a diabetes risk questionnaire based on the same risk factors as in Denmark. Those at high risk of having type 2 diabetes are requested to come for a screening visit at a centre set up near the general practice. A stepwise screening strategy based on RBG, FPG and OGTT is used to diagnose diabetes In UK, different strategies are used in Cambridgeshire and Leicester. In Cambridgeshire a search of computerised general practice records is performed, using a simple validated risk score, based on routine general practice data (age, gender, prescribed medication and body mass index), to identify people in the age of 40-69 years at high risk of having undiagnosed diabetes30. Those with a high score undergo a stepwise screening strategy based on RBG, FPG and OGTT is used to diagnose diabetes In Leicester all white European subjects aged 40-75 years and Asian, black or Chinese subjects aged 25-75 years are invited in a restricted geographical region within Leicester. All attendant undergo an OGTT as the first screening step (unless FPG > 7.0 mmol/l)

Diagnostic procedures:

The diagnostic procedure includes a stepwise procedure minimizing the work-load on the general practitioner and includes the questionnaire, random capillary blood glucose and HbA1c as screening instruments followed by fasting capillary blood glucose (FCBG) and an oral glucose tolerance test in everyone with marginally elevated FCBG. The diagnostic criteria follow the most recent World Health Organization guidelines (31) and are based on two diagnostic glucose values measured on independent days.

Exclusion criteria:

Participants are excluded if they already have diabetes, are pregnant or lactating or have a severe psychotic illness, are house bound or have an illness with a likely survival of less than one year.

Outcome measures:

Outcome measures from the screening study include measures of the efficacy of the screening campaign, the objective health status of patients newly identified by the campaign, feasibility as reported by the primary care physician, and the economic impact or benefit of the a programme. Furthermore a substudy explores the psycho-social and ethical aspects of the screening programme.

Methodology - Intervention study:

All patients diagnosed as part of the screening programme are invited to enter the ADDITION-study. The study will include a minimum of 3000 patients with screen detected diabetes. The general practices are randomised to the routine care group (standard care as given by the GP) or to the intervention group which features a target driven, intensive multifactorial approach to treatment. The study is an open, multicentre, parallel group trial with randomisation of general practices. Patients are screened and recruited during the period January 1st 2000 and June 30th 2006. The end of follow up is by July 1st 2009. Participation is based on informed consent in accordance with the Declaration of Helsinki.

Intensive Treatment strategy:

The intensive multifactorial treatment includes lifestyle advices (concerning diet, physical activity, medication adherence and tobacco cessation), prescription of aspirin and stepwise increases in pharmacological treatment of blood glucose, blood pressure and lipids, according to strict targets (appendix 1). The treatment targets are as follows:

• HbA1c < 7.0
• Total cholesterol < 5.0 mmol/l (4.5 mmol/l if CVD present)but statin to everyone with total cholesterol > 3.5 mmol/l
• Blood pressure < 135/85
• Aspirin 75- 150 mg/day to everyone on antihypertensive treatment Behavioural advices include
• Smoking cessation
• Physical activity 30 min./day
• Healthy diet (low fat, 600g of fruit and vegetables/day)

The treatment targets may be intensified during the study according to the results of other clinical trials published during the study period, as the aim is to strive for treatment targets based on the most intensive guidelines available.

Within the intensive group a further randomisation allocates 50% of the patients to country specific interventions concerned with improving adherence to lifestyle changes and medication. This intervention, including the use of motivational interviewing) is delivered either by a trained facilitator (UK and The Netherlands) or through training of practitioners (Denmark), and is based on a client-centred non-directive counseling style to help patients explore and resolve ambivalence and stimulate lifestyle changes, appropriate diabetes self-care and adherence to medical treatment 32,33.

Pharmacological treatment:

The decision on which pharmaceutical drug to use for the individual patients with is made by the clinician as the study is target driven and not a trial comparing different specific drugs. The clinician is provided with recommendations for a treatment strategy (appendix 1), which should be based on balancing treatment effect, side-effects and cost. The main priority is achievement of treatment targets with a flexible lifestyle and low rates of side effects such as hypoglycaemia and weight gain.

Therapies are adjusted at 2 to 4-weekly intervals until targets are reached, thereafter every 3 months. HbA1c is taken every third month, in between antidiabetic drugs are adjusted according to blood glucose measurements in the interim.

Outcome measures: see other section.

Sample size and statistical power:

Based on levels of risk in the conservative-treatment arm of the UKPDS, the expected event rate is 3% per year for the combined endpoint (all-cause mortality, nonfatal myocardial infarction, stroke, revascularisation or amputation). With a sample size of 1350 patients in each arm (standard versus intensive treatment) the study will allow the detection of a 30% risk reduction in the intervention group at a significance level of 5% with a mean duration of follow-up of 5 years.

Timescale:

The screening study began in late 2000 and will end by June 30th 2006. Patients are enrolled into the treatment study following diagnosis in the screening study: thus the treatment study proceeds along with the evaluation phase of the screening study. The follow-up will continue until July 1st 2009.

Ethics and safety:

The Scientific Ethics Committee in the involved counties in Denmark and the Multipractice Study Committee have accepted the project, and the study has been approved by the ethical committees in UK and the Netherlands.

A data safety and monitoring committee will have access to all end point data (unblinded) after 1, 3 and 5 years, and the study will be terminated if the composite end point (including: Cardiovascular death, non-fatal MI, non-fatal stroke, revascularisation and amputation) should demonstrate a clear advantage of intensive versus standard treatment (p<0.001) or a clear advantage of the standard versus intensive treatment (p<0.01).

Perspectives of the trial:

The results of the study will be of immediate national and international relevance to policy decisions about screening for diabetes, and subsequent intensive treatment. If the study shows that screening and early intervention markedly reduces the risk of developing premature CVD, then the study will have potential important impact at the individual patient level as well as on the societal level. If the study fails to show an effect of screening and early intensive treatment then this would have direct impact on future recommendations for screening and treatment regimens, and the study could lead to significant savings if it shows that strict treatment targets are not essential in the early phase of the disease. The results of the sub-study will inform approaches to health promotion to the management of chronic disease and risk, and to strategies to support adherence applicable not only to diabetes but also to other chronic diseases.

References and Appendix 1 on www.addition.au.dk