Last week as I towelled down the patient, mopped the surgery floor and hung my jacket up to dry I realised that it was a long time since I had syringed ears.

Fifteen years ago as a singlehanded GP, I also took blood, gave injections, did night visits and shared maternity care. Most of these tasks are now being done by others.

Despite my feelings of nostalgia, I recognise the need to reform some aspects of primary care. I acknowledge that there are strong arguments in favour of nurses taking a lead role in chronic-disease management and for pharmacists to take on medicines management.

I would also be delighted if more dentists could be employed by my local primary care trust to remove some of the dental patients from my emergency list. But suggesting that primary care professionals are a homogenous group and that we need to 'trade down' to the cheapest option is foolish.

In the drive to remove clinical tasks from the GP, PCTs must avoid equating common problems with unimportant problems. I have come across instances of patients dying from their 'unimportant' low back pain either acutely due to a ruptured aortic aneurysm or more slowly from widespread prostate cancer. All common symptoms can, on occasions, indicate a serious underlying condition and it takes considerable experience to separate the wheat from the chaff.

Some patients may also be missing out on the 'door handle' moment when they turn back to you and explain the real reason for the consultation. I recall the patient with testicular cancer who came in with a sore throat and the mother with breast cancer who brought in her child with mild eczema. I worry that nowadays with PCT-initiated 'minor conditions' local enhanced services, the outcomes for such patients might have been very different.

Recently a consultant cardiologist lamented to me the loss of clinical assessment skills by junior doctors.

In his hospital, nursing staff extract the clinical information from the patient and the junior doctor simply copies it, focusing their energies on selecting the most appropriate investigations. This, together with an inadequate clinical assessment in general practice, is dangerous.

Weight loss is another common symptom, with a range of possible causes including 'organic' diseases such as cancer and thyroid dysfunction as well as 'non-organic' problems such as anxiety and depression. Recently I saw a patient with weight loss who had been investigated for this symptom without any GP input.

When I eventually met her she was obviously depressed and the question I asked was simple: 'Did you feel depressed before you went through all the tests?' She burst into tears. 'I have always felt low but no-one asked.' All her investigations were normal and she is now being treated for depression.

Over the past 15 years nobody has ever asked me if I no longer wish to see so-called minor conditions and, as far as I am aware, no patients have ever been consulted either. It seems that decision making in this area is often based on personal experience and anecdote rather than hard evidence. More than once I have heard managers and policy makers claiming to know all about general practice because they have a relative who is a GP.

My grandfather was a plumber, but I would be hard pressed to convince the surgery staff that the puddles on the surgery floor are due to leaky pipework. l Nick Summerton is a GP and reader in public health and primary care at Hull University.