DTU Logo

UK Prospective Diabetes Study

Main trial

Type 2 Diabetes - a common and dangerous condition

Why UK Prospective Diabetes Study was needed and why its results are important

Professor Robert Turner, Professor Rury Holman, Dr Charles Fox, Dr David Wright, Professor David Hadden


There are two common types of diabetes. Type 1, which usually starts in young people and always needs to be treated with insulin. Type 2 diabetes, that occurs in 90% of people with diabetes, starts later in life and can often be treated with diet and/or tablets. Although apparently a "mild form of diabetes" major complications occur, including death from heart attacks and disability including blindness, amputations and kidney failure. UK Prospective Diabetes Study (UKPDS) was a unique, long-term clinical trial designed to find out how to treat people with Type 2 diabetes to prevent complications and to maintain their health.

UKPDS asked simple questions:

The study has been successful and has shown that the complications of diabetes can be prevented by:

These two approaches have now been shown to be the key to maintaining the health of patients. Improving blood pressure is sometimes easier than improving blood glucose levels, and needs to have a high priority.

Now it is known that the complications can be prevented with more intensive management:


Type 2 Diabetes

Diabetes is the name given to a disease in which high blood glucose concentrations occur. Type 2 diabetes is a common condition, affecting 2% of the population and 8% of people in their sixties. People with diabetes usually present with thirst, weight loss and tiredness. In people with Type 2 diabetes that is usually diagnosed in middle age, symptoms can often be controlled by diet or tablet therapy. It is therefore frequently called "mild diabetes". However, glucose levels usually remain high and these can lead to "diabetic complications". Type 1 diabetes is the form of diabetes that usually comes on in childhood or in early adulthood and requires life-long insulin treatment. Type 2 diabetes accounts for approximately 90% of diabetes, Type 1 diabetes approximately 8% , with 2% from other causes.

Complications of diabetes from large and small vessel damage

The "complications" can arise from damage to large arteries that cause heart attacks and strokes. These occur two or three times more often in people with Type 2 diabetes than in the general population and can cause death at an earlier age than is usual. If not fatal they can seriously affect the quality of life. Specific diabetic complications can also develop, due to closure of small blood vessels that can cause blindness, kidney failure and amputations. These kinds of complications are also found in people with Type 1 diabetes.

Lack of knowledge of how to prevent diabetic complications

Until now there has been little convincing evidence that improved diabetes control will prevent the complications of diabetes. In addition, the few clinical trials that have been done have raised the possibility that some of the commonly used treatments may aggravate damage to large arteries. Because most people with Type 2 diabetes do not have symptoms, they are often not intensively treated, and may visit their physician only once or twice a year for a general check. This has been an acceptable clinical practice, in view of the lack of reliable evidence to show that intensive treatment of blood glucose or blood pressure will prevent complications, and thus help to maintain the health of patients.

Why has there not been better evidence to guide physicians on how to treat people with Type 2 diabetes?

The problem has simply been that the complications of diabetes come on over decades. Therefore a study to show whether a treatment will prevent complications has to be a long-term study. It also needs to be on a large scale, studying large numbers of patients, to obtain a reliable answer. UKPDS was a unique, long-term study. It was planned in 1976, and recruited over 5000 newly diagnosed patients in 23 sites in the UK (2 in Scotland, 2 in Northern Ireland and 19 in England). The study finished at the end of 1997.

Was UK Prospective Diabetes Study really necessary?

Yes, very necessary. In 1993 a large American study (DCCT) was published which showed a close relationship between control of Type 1 diabetes and the risk of complications from small vessel disease. In Type 2 diabetes diabetes, the major problem is complications from large artery disease and it has not been known whether more intensive therapy would be helpful, particularly as there was concern about whether in the long term some therapies could themselves cause large vessel disease. Thus doctors treating the most common form of diabetes did not know how hard they and people with Type 2 diabetes should strive to keep blood glucose near-to-normal. In addition, no larger scale studies had been done to show whether improved blood pressure control was effective.

Confirmation that Type 2 diabetes is dangerous

UKPDS, sadly, confirmed that patients with Type 2 diabetes have a high incidence of heart attacks and strokes, and have a greater likelihood of early death than the general population. After 10 years of diabetes, the people with diabetes in the study had a 2-fold greater mortality than the general population. The study has also shown that after 10 years one third have a complication that requires clinical attention, including heart attacks, strokes, laser treatment of the eyes, treatment for renal failure or amputations. Patients also develop cataracts requiring surgery.


1. First question of UKPDS: Will patients benefit from better blood glucose control?

The main questions asked were:

  1. Will a policy of intensive blood glucose management prevent complications?
  2. Is any one treatment better than others?
  3. Is any one treatment contra-indicated?
  4. For what level of blood glucose control should one aim?

The study was set up so that physicians could have a sound basis for deciding which treatments they should prescribe to patients.

An intensive blood glucose control policy decreased the risk of diabetic complication

The patients were randomly allocated to different therapies, diet, sulphonylurea tablets, metformin tablets or to insulin.

UKPDS showed conclusively that in people with improved blood glucose control the risk was reduced of:

In addition, there was some evidence, albeit inconclusive, that:

The therapies studied were each shown to be effective

2. Second question of UKPDS: Will patients benefit from better blood pressure

Early in the course of the Study, it was apparent that many patients had high blood pressure and they were particularly prone to heart attacks and other complications of diabetes. Therefore in 1987, an additional study was set up in over 1100 people with diabetes and a high blood pressure. The main questions were:

  1. Will a policy of intensive blood pressure control prevent complication?
  2. Is one treatment better than the others? (There have been suggestive, but not conclusive evidence, that one type of drug, ACE inhibitors, may have advantages)
  3. Is any one treatment contra-indicated?

The tight blood pressure control policy decreased the risk of diabetic complications

UKPDS showed that improved blood pressure control decreased the risk of:

ACE inhibitors and beta-blockers were each shown to be effective.

Management Questions

If intensive management of high blood glucose and high blood pressure are beneficial, what advantage is gained from better treatment or both?

The Study was not primarily designed to answer this question. Nevertheless, the available data suggest that if more care is given to obtain both near-normal blood glucose and near-normal blood pressure levels, the benefit from each therapy would combine to give a greater reduced risk of complications than either treatment on its own.

When should intensive management of diabetes start?

As soon as diabetes is diagnosed, to give best chance of preventing complications. UKPDS has shown that, by the time diabetes is diagnosed, approximately half of people with Type 2 diabetes already have evidence of diabetic tissue damage. The results of UKPDS suggest that a formal screening programme should be set up to identify diabetes before symptoms occur. This would allow early therapy to be given before the processes which lead to complications can operate.

Does intensive management of diabetes affect quality of life?

Questionnaires were given to people in the study to assess their quality of life with questions assessing mood, work satisfaction, symptoms and everyday mistakes. These showed that intensive management was neutral in its effect, with neither an adverse or beneficial effect. On the other hand, the questionnaire showed that when people had diabetic complications, these adversely affected quality of life.

Is the intensive management costly? Will it be cost-effective?

The increased cost of medication for improving blood glucose control or blood pressure control was directly recouped by the lower costs from fewer clinical complications. The nursing and physician costs of the more intensive management were not recouped. Nevertheless, in comparison with other preventative strategies, the intensive management could be regarded as being cost-effective.

Other conclusions

Improved blood glucose control can be obtained

The study showed that intensive blood glucose treatment is acceptable and that better blood glucose control could be obtained than with conventional management. Nevertheless, the study showed that Type 2 diabetes is a progressive disease, and with time more treatment is needed. People with diabetes may often require three different types of tablet and even then insulin treatment is required by many patients. Some of the treatments, including insulin, had the downside of giving the risk of 'hypos' -- episodes of low blood glucose -- and of greater weight gain of approximately 8 lbs (4 kg). The Study has shown that insulin is an acceptable treatment when given at an earlier stage of Type 2 diabetes than currently is the practice. In spite of insulin therapy, after a few years it can be difficult to maintain good blood glucose control. New methods of treating diabetes are still required.

Improved blood pressure control can be obtained

The study showed that intensive blood pressure treatment is acceptable to people with diabetes, and that better blood pressure control than standard management can be obtained. Within nine years approximately one-third of patients required three different types of treatment. Serious side -effects were unusual.

Which is more important?

Both are important, however, obtaining near-normal blood pressure is, in many patients, easier than maintaining near-normal blood glucose levels because more different types of therapies are available. Since medicine is the art of the feasible, a high priority could be given to ensuring good control of blood pressure.

Which glucose or blood pressure levels should be attained?

Different goals apply according to the age and type of diabetes.

The future?

At present diabetic complications are often regarded as being a natural outcome from a chronic disease. The study shows that this is not the case and that with good management, including improved blood pressure and blood glucose control, may help to make complications less common and no longer a half-expected outcome.

Additional Information About UKPDS


In Type 1 diabetes a large American study (DCCT) showed that there is a close relationship between the control of diabetes and the risk of developing certain specific complications, such as eye, kidney or nerve disease. However, in Type 2 diabetes the major problem is an increased likelihood of developing heart attacks, and it is not known if any treatment will prevent these.


The UKPDS is the largest and most comprehensive study of diabetes that has been undertaken. The UKPDS was set up in 1977, initially as a pilot scheme involving five sites in England, Scotland and Northern Ireland. The study was subsequently extended to 23 sites. In 1987, the study of treatment of hypertension was started. UKPDS recruited 5100 patients who were allocated to various treatment groups. From the point of view of their age and racial mix, these people were representative of the typical British patient with Type 2 diabetes. UKPDS closed at the end of 1997 and the results published in September 1998.

Structure of the UKPDS

After an initial treatment with diet alone for three months, people with a raised fasting blood glucose were randomly allocated to one of various treatment policies. Some people were allocated to a conventional policy, primarily with diet alone. Moderately high blood glucose levels were accepted into this group. Others were allocated to an intensive policy group aiming for near-normal glucose control, with two different sulphonylureas tablets or to insulin. Overweight people could also be randomised to a tablet called metformin, as an additional option. Each of these treatments has now been shown to be advantageous. People with raised blood pressure as well as diabetes were randomly allocated to a policy aiming for tight blood pressure control, using two treatments that might be thought to be particularly beneficial, an ACE inhibitor or beta-blocker or to a less tight blood pressure control. Both ACE inhibitors and beta-blockers have now been shown to be advantageous.

Post Trial Monitoring

When the intervention phase of the UKPDS finished in 1997, all surviving UKPDS patients were entered into a ten-year, post-trial monitoring program. In 1998, following presentation of the trial’s primary results at the EASD meeting in Barcelona, patients and clinicians were advised of the need to lower both blood glucose and blood pressure levels as far as feasible. Patients returned to community or hospital-based diabetes care, according to their clinical needs, but with no attempt to maintain previously randomized therapies. They were seen annually until 2002 in UKPDS clinics with continued standardized collection of endpoint, biochemical and clinical data. Any patients unable to attend clinics were sent EQ-5D and health resource questionnaires, with additional questionnaires to their general practitioners to captured possible endpoints. These same questionnaires were used to follow all patients remotely from 2002 to 2007, as financial constraints meant UKPDS clinics could no longer be supported. The same endpoint ascertainment and adjudication process was used throughout to ensure analytic comparability.

The purpose of UKPDS post-trial monitoring was to:

  1. Observe the glycaemic and blood pressure control achieved as patients returned to their usual care physicians.
  2. Monitor the glucose and blood pressure lowering therapies prescribed.
  3. Look for possible “legacy effects” of earlier improved blood glucose and/or blood pressure control.
  4. Examine risk factor associations and the possible influence of randomised therapies on less frequent clinical and biochemical outcomes as more events were accrued.