JANE LISTER remembers the day she started to feel pains in her shoulder. She had been sitting in the driving seat of her car and had leaned back to open the rear door for one of her children. She ignored the discomfort, but about a month later her shoulder became increasingly stiff and painful.

Mrs Lister, 48, began waking at night with the pain. She found it difficult to change gear when driving and to lift items off the kitchen shelves. Worried that arthritis had set in, she went to her GP. He found that she was unable to turn her arm outwards at the shoulder and had lost the normal rotation of the joint.

The doctor told Mrs Lister that she had a 'frozen' shoulder. But it was unlikely that it had been caused by the reaching into the back of the car; more likely, her shoulder had already been stiff - but without pain - and she had then stretched it. He recommended an intra-articular (into the joint) steroid injection and a course of home exercises.

As many as four in 100 people will suffer a 'frozen' shoulder (or adhesive capsulitis, as it is clinically known) at some time. It seems to be an increasingly common problem in all age groups, but specialists are at a loss to explain its exact cause. Frozen shoulder hardly ever results in an operation: this usually is necessary only as a result of long-term arthritis or inflammatory rheumatoid arthritis.

Frozen shoulder, first identified in 1896 by a French physician, Duplay, is best described as a spontaneous onset of shoulder pain - in the absence of any other bodily illness - accompanied by increasingly severe limitation of movement. One possible reason for it is that the collagen, the main component in connective tissue in tendons, grows weaker with age; a few minor awkward movements can cause stiffness and pain.

'To some extent, it is one of the effects of the ageing process, but not all cases of frozen shoulder can be put down to this reason,' says Professor William Wallace, consultant orthopaedic surgeon at the Queen's Medical Centre at Nottingham University.

Patients with osteoarthritis or rheumatoid arthritis are usually more susceptible to the condition. Otherwise it tends to occur in older people; the most common age range of sufferers is 40-50. Shoulder joints become less flexible with age as they are used less in physical exercise. In some cases this can lead to restricted ability to rotate the shoulder. Twice as many women as men are sufferers.

In people under 40, the main problem is instability of the shoulder joints, which become easily dislocated. People who are active in sport or other activities where the shoulder muscles are used may be more susceptible, Professor Wallace says. In older people, the problem is caused by tears in the rotator cuff, the part of the shoulder joint that controls rotating motion.

Frozen shoulder can be extremely painful. It has three phases that can last anything between 12 and 24 months: the 'freezing' stage, usually the most painful, lasts for about six months; the 'frozen' stage, when stiffness increases, from three to six months; and the 'thawing' stage from six to twelve months.

In a study of 47 patients, published in the British Medical Journal last year, Professor Wallace and his colleagues found that intra-articular steroid injections into the joint brought an improvement in nearly all patients. During the 'freezing' stage, most sufferers consult their GP, and are sometimes offered a pain-relieving steroid injection. But many GPs have been reluctant to give the injection because it is a difficult procedure and they were not trained to adminster it.

This is not much comfort to the sufferer, who may have to wait a long time, in pain, for referral to hospital (patients on Professor Wallace's waiting list are having to wait a year).

But a new development in treating painful shoulders may encourage more GPs to give the injection. Colin Howie, an orthopaedic surgeon at Raigmore Hospital in Inverness, has pioneered a life- like model shoulder, complete with artificial skin, that allows GPs to practise the injection technique without the need for a human guinea-pig. (Until now, the only way a doctor could gain experience of giving the injection was by using volunteer medical students or patients.)

'A lot of GPs used to come up to the hospital to practise on my patients with shoulder problems, and I didn't like that,' says Mr Howie. 'So I thought I would develop a model for GPs to practise on. When we carried out a trial with the pilot model, we found that GPs, who had not formerly given the injection, doubled their confidence, and those who had given it before became slightly less confident that they had been doing it correctly.'

The latest version of the model, developed with advice from the Spitting Image team, is being supplied by E Merck Pharmaceuticals. The shoulder, made from a synthetic, self-sealing skin-like material, contains a simulation of sinew and cartilage, with sensitive electronic pads inserted at the four most common sites for the steroid injection.

The needle should either be inserted at the head of the humerus, the bone of the upper arm, or between the head and the glenoid, the shoulder joint. If this is done correctly, the end of the demonstration syringe lights up. If the doctor 'misses' because the inclination or position of the needle in the shoulder is not correct, he or she must try again.

Giving the injection properly takes practice. In a teaching session at Nottingham, only two of five GPs succeeded, even after two or three attempts. But GPs may find it worthwhile to learn how to carry it out: under their contracts, introduced in 1990, they are paid for minor surgery.

(Photograph omitted)