People with Type 2 diabetes should not be encouraged to monitor their own blood sugar levels, research shows.

The NHS is estimated to spend more than £100m a year on helping those who do not rely on insulin to test themselves.

But two studies published in the British Medical Journal suggest this group is not reaping any long-term health benefits and could be making themselves depressed.

One study assessed if self-monitoring had an effect on blood glucose levels and the incidence of hypoglycaemia in 184 people newly diagnosed with Type 2.

The researchers at the Western Health and Social Care Trust in Londonderry found self-monitoring had no significant effect after a year.

But it found that self-testing was associated with a 6% higher score on a depression and anxiety scale.

A second study examined 453 patients with Type 2 diabetes who did not rely on insulin.

It found that self-monitoring was linked to higher costs and lower quality of life for patients.

The Oxford University study found no convincing evidence for recommending self-monitoring and found that the additional healthcare costs were about £90 per patient each year.

About 1.8 million people in the UK have Type 2 diabetes and another 750,000 are thought to be undiagnosed.

With diabetes now affecting 3-4% of the population of the UK, the total healthcare cost of self-monitoring may now exceed £100m each year.

Professor Martin Gulliford, of King's College London, said more attention should be placed on monitoring blood pressure, cholesterol and body weight.

Libby Dowling, of Diabetes UK, said more than 95% of diabetes care is self-management.

Self-monitoring must be accompanied by a structured education programme to teach people when to test, she said.

"Poorly controlled diabetes can increase the risk of complications such as heart disease, so short-term savings made by reducing self-monitoring could be dangerous," Ms Dowling warned.