GPs are once again being urged to reduce the number of antibiotics they prescribe amid concerns that overuse is wasting resources and increasing antibiotic resistance. But it is not just doctors like me that are being targeted in the latest initiative — patient pressure for antibiotics is a major influence on the prescribing habits of doctors, and it is going to take a significant change in behaviour on both sides of the surgery desk if we are to make a difference.

The Department of Health launched a major advertising campaign earlier this year to remind people that minor, self-limiting viral infections — including most coughs and colds — do not respond to antibiotics. The National Institute for Health and Clinical Excellence (NICE) has issued new guidance for doctors suggesting that antibiotics should no longer be used routinely in other upper respiratory conditions such as sinusitis and tonsillitis.

The net result should be a dramatic reduction in the number of prescriptions issued, and there is room for improvement: prescribing rates have fallen over the past decade, but the average British man can still expect around 50 courses of antibiotics during his lifetime, a woman closer to 70. Five million prescriptions for antibiotics were issued to children last year.

Book in to see your GP with a cold, cough, bronchitis, sore throat, tonsillitis, earache or sinusitis this winter and you are likely to be sent away with nothing more than advice on do-it-yourself symptom relief. This group of respiratory tract infections accounts for more than half of antibiotic usage in general practice, yet most are caused by viruses, which are resistant to all antibiotics.

Issuing a prescription for these self-limiting conditions will have no impact on the duration or severity of the illness and may actually cause more problems resulting from side-effects such as allergic reactions, diarrhoea and the wiping out of the billions of healthy bacteria living in the gut.

There are wider implications, too. Over-dependence on antibiotics fuels the emergence of resistant “superbugs”, such as MRSA, and encourages people to see a doctor for a self-limiting infection that they are unable to influence, wasting valuable time and resources for all concerned.

A quarter of the British population — 15 million people — visit their GP at least once every year with a respiratory tract infection. I saw three patients with colds on Friday, and I am still not sure why they thought that they needed to see me or what they were expecting me to do for them.

But not all respiratory tract infections are viral and self-limiting — some are caused by bacteria, and can turn very nasty if antibiotics are not prescribed. Herein lies the major hurdle for both doctor and patient. How do you tell the difference?

It is all very well for researchers to comb through published studies showing that antibiotics have no impact on the severity and duration of symptoms in the average patient with a cough, but in general practice there is no such thing as an average patient.

In nine out of ten cases antibiotics won’t make any difference, but for the remaining patient they could mean the difference between a troublesome cough and life-threatening pneumonia.

The new NICE guidance has tried to address this uncertainty by outlining when doctors should have a lower threshold for prescribing antibiotics while at the same time providing patients with a useful insight as to when they should see a doctor. The full guidance can be found at www.nice.org.uk , but here are some of the key features:

• Earache in children. Antibiotics should be considered for bilateral ear infections in children under two, and in children of any age with a discharging ear. Other cases can generally be treated with pain relief only.

• Sore throat / tonsillitis. Use antibiotics only in patients with a sore throat and three or more of the following: pus on the tonsils, tender lymph glands in the neck, no cough, a raised temperature.

• Cough. In the absence of signs and symptoms suggestive of more serious infection (such as shortness of breath, breathing faster than normal or coughing up blood), antibiotics are best avoided. The duration of a cough is not an indication of the need for treatment either — a typical viral cough can easily take two to three weeks to settle. Coughing up yellow/green phlegm is to be expected.

• Notable exceptions include the elderly and patients at risk because of underlying problems such as heart and lung disease or a suppressed immune system, in whom there should be a much lower threshold for resorting to antibiotics.