Not a reassuring white-coated boffin - more likely a teenage school- leaver paid around pounds 8,000 a year... Jeremy Laurance reports on the shocki ng crisis in the screening service
On the kitchen table stands a box of slides and a microscope. Between picking up the kids from school and slicing the carrots for supper, amid the debris of half-eaten Marmite sandwiches and cut-out-and-paste models, the screener goes about her business of examining the slides for signs of cervical cancer.

Most people will find it shocking that a test as critical as the detection of cancer could be carried out in the kitchen of somebody's home. To them, the NHS means white-coated, highly qualified staff, pounds 1 million body scanners and expensive drugs, not a home-spun service where smears are checked while the dinner is cooking and the kids are watching television. Yet the cervical cancer-screening programme has always been a Cinderella service, dependent on the lowest tech facility - the poorly-paid human screener. Most are women, many still in their teens, who have a handful of GCSEs, a couple of years training and are paid about the same as a daily cleaner.

The low pay - around pounds 7,000 to start, rising to a maximum of pounds 11,000 - and the tedious, difficult work have made screeners hard to recruit, which is why some laboratories have allowed them to work from home. The practice of home screening is now ending, after it was exposed last year in one of the many scandals that have beset the cervical cancer screening service. A woman in Gorleston, Norfolk, who worked at home for the James Paget Hospital, made a series of errors that led to 37 women with moderate to severe abnormalities being misdiagnosed. In all, she misread 357 smears out of 10,000 that passed through her hands over three years.

The NHS Cervical Screening Programme declared itself horrified at the disclosure that screeners were working from home and ordered the 14 laboratories still using them to stop the practice. However, since it takes two years to train a screener, the programme was forced to allow the laboratories affected two years grace to find and train replacements. Some will therefore still be dependent on the kitchen-table checkers until next year.

The episode highlights the difficulties faced by a service that began in a haphazard way 30 years ago, and is now being updated to meet the standards of the 21st century. Cervical screening started in a few local areas in the early Sixties as a result of the enthusiasm of a handful of committed gynaecologists. As it grew, each area developed its own way of doing things. In 1988, a national scheme was established to encourage all women of reproductive age to have a regular smear, and in 1994, because of concern about the quality of screening, management was transferred to the NHS breast screening programme.

This month, Frank Dobson, the Health Secretary, announced a review of the breast and cervical screening services which are to be required to meet standards laid down by the health department. A major drive has since been introduced to improve the training of screeners, bolster management and plug holes in the system to ensure abnormal smears do not fall through the net. The changes have won praise from those at the top but there has been resistance at grass-roots level. One manager says: "We are up against people who have been doing things the same way for 30 years and don't see why they should change."

The need for change is demonstrated by recent events. Earlier this month, Warwickshire Health Authority announced it was re-checking 18,000 smears after monitoring revealed some had been misread. It blamed the errors on the inexperience of the screeners at the laboratory at St Cross Hospital, Rugby, which was processing 7,000 smears a year - too few to provide an adequate mix of normal and abnormal smears. This problem was recognised a decade ago, yet despite guidance from the health department, setting a minimum 15,000 smears for each laboratory - the latest issued last year - 33 labs are still processing fewer than the minimum.

There is also an unexplained variation in the proportion of smears judged abnormal in different parts of the country, ranging from less than two per cent to more than 10 per cent. This five-fold difference cannot be accounted for by differences in the incidence of the cancer and means that some women with early signs of the disease are being missed while others are being needlessly worried. Laboratories have been told that they should expect to find between four and seven per cent of smears with mild abnormalities and up to two per cent with moderate or severe abnormalities but many fall outside that range.

Errors in screening can creep in at any of a dozen points in the process. In taking the smear, the circumference of the cervix must be scraped as cancerous changes can begin at any point on its rim. (They may also begin inside the inaccessible endocervical canal, which is why some cancerous changes will always be missed.) The smear must be fixed by soaking it in alcohol, labelled and sent to the laboratory where it is numbered and stained. The primary screeners, who need a minimum of four GCSEs and undergo two years training, spend about eight minutes on each slide and have to spot as few as two or three altered cells among the 300,000 to 500,000 present in the average smear. They work a maximum of four hours a day in two shifts of two hours on screening, spending the rest of the day on other work.

Because the work is difficult and tedious, and the screeners tend to be young and poorly paid, there is an elaborate checking process. One consultant cytopathologist says: "Everyone makes mistakes. You come in with a headache, you are screening smear after smear and you may miss a few. That's why we have these checks." Negative slides, judged normal, are re-screened in a quicker process taking two minutes before being cleared. Abnormal slides are checked by a senior technician before being passed to the consultant pathologist for a final decision.

The shortage of screeners is a major problem. Consultant cytopathologists (specialists in cervical screening) who run the programme say improvements cannot be achieved unless the pay of the screeners is increased. Dr Lesley Turnbull, secretary of the British Society for Cervical Screening, says: "Screeners do an extremely demanding job for minimal pay. Now they are feeling got at. Everyone is saying, 'Look at what you missed.'"

Worse than that is the shortage of consultant cytopathologists of whom there are only about a dozen in the country. Most screening labs are run by general consultant pathologists with other interests, who cannot give screening the attention it requires. Dr Turnbull says: "The main weakness is at the consultant end. A lot of pathologists don't have the time to commit to cytology (checking smears). You cannot do it for half an hour a week because you will never develop the necessary expertise."

Despite the problems, figures show the screeners are doing a worthwhile job. The screening programme is estimated to be saving up to 4,000 cases of invasive cancer a year, though it fails to detect it in a further 1,500 to 2,000. Cases of cervical cancer have declined 30 per cent in a decade, from 16 to 11.8 per 100,000, and deaths have fallen by more than 40 per cent since 1979, from seven per 100,000 to 4.1 per 100,000. Half of the 3,500 cases of invasive cancer occurring each year are in women who have never been screened.

Julietta Patnick, co-ordinator of the national programme, says it is essential that women do not lose faith in the screening service. "We can't excuse what has gone on, but the mortality and incidence are dropping like stones. The programme is working well but it could be better."