MJA: Wain, Automation in cervical cytology: whose cost and whose benefit?

Automation in cervical cytology: whose cost and whose benefit?

Our first priority should be to screen more women, and not to offer more screening to those who can pay

MJA 1997; 167: 460-461

Readers may print a single copy for personal use. No further reproduction or distribution of the articles should proceed without the permission of the publisher. For permission, contact the Australasian Medical Publishing Company
Journalists are welcome to write news stories based on what they read here, but should acknowledge their source as "an article published on the Internet by The Medical Journal of Australia <http://www.mja.com.au/>".

Make a comment - Register to be notified of new articles by e-mail - Current contents list - ©MJA1997


	Recently, women and doctors have been exposed to high pressure marketing about automated cervical cytology. Questions as to whether the patient is "getting the right Pap smear", and messages that doctors don't know what is being missed have created a climate of uncertainty and anxiety among both patients and practitioners. Population screening for cervical cancer by Pap smear has undoubtedly had a major impact on both morbidity and mortality from cervical cancer,¹ a malignant disease which, on a worldwide basis, is the commonest cause of cancer death in women. The success of cervical screening has raised expectations for a perfect system and has also directed focus to its failures. Failures can occur at any point of the screening pathway: at recruitment, sampling, laboratory processing, notification, and in the management of women in whom abnormalities are detected.² While systematic efforts have been made to improve each step of the pathway, the major failure remains the inability to persuade all women to undergo regular Pap smears: figures from the NSW Pap Test Register suggest that only 67% of women at risk of cervical cancer are screened in accordance with the national policy of biennial Pap smears.³ The remainder are failing to take advantage of the well established health benefits of regular conventional Pap smear screening.
The current marketing strategy runs the risk of reducing confidence in the existing system and of introducing apparent inequity into cervical screening	Attempts to reduce laboratory error in the processing of Pap smears have encouraged the automation of cytology, with developments on three fronts. Firstly, the use of liquid-based smear preparation which prepares a monolayer of cells on a slide in the laboratory; secondly, the use of microscope tracking to ensure that the cytotechnologist examines all areas of a slide; and thirdly, computer-assisted image analysis to rescreen conventionally prepared smears. All procedures ultimately rely on conventional cytopathology review and "they still do not detect all abnormal cases".⁴ Some of these techniques have been heavily marketed, both directly to women and to general practitioners, and none currently carry a Medicare rebate. The role of automated cervical cytology in Australia is currently being evaluated by an Australian Health Technology Advisory Committee Working Party and its report is expected shortly. This Journal has previously reported the experience of one Australian laboratory with a computerised rescreening technique, PapNet,⁵ and, in this issue Roberts and colleagues report the findings of another large laboratory on the use of a liquid-based technology, ThinPrep, as an adjunct to the conventional Pap smear.⁶ Both papers suggest a slight improvement in detection of abnormalities. The medical community must remain cautious in assessing the true place of such technology. These techniques have been marketed as offering improvements in laboratory quality assurance and thereby reducing false negatives. Additional claims for the liquid-based techniques (which prepare better slides) are that sampling is improved and the number of unsatisfactory samples reduced. They also theoretically open the way for additional tests such as typing for certain strains of human papilloma virus (HPV). As with all of the automated strategies, the impact on clinical outcomes of HPV testing still remains conjectural.
	The performance of automated techniques in quality assurance should be assessed against other methods of quality assurance, such as random rescreening of a mandated proportion of smears, directed rescreening of "high-risk" groups and "rapid rescreening". Mathematical models show that manual methods of rescreening provide superior cost-benefit ratios when compared with automated approaches.⁷ This conclusion was also reached by the Canadian Coordinating Office for Health Technology Assessment in its assessment of automated cytology,⁸ and is supported in a letter in this issue of the Journal by Gurley et al.⁹ Pathologists should be encouraged to establish the best and most cost-effective method of quality assurance and not simply pass on the cost of the most expensive method of quality assurance to consumers. The scientific assessment of these procedures has been clouded by opinions about potential medicolegal consequences for failure to recommend the use of such tests.¹⁰ Such suggestions fail to recognise that proof of medical negligence must involve a deviation from the standard of care, which is ultimately decided by the courts, but depends on the ordinary care and skill of a particular category of practitioner. Australian medical negligence cases involving cervical cancer have focused on the failure of doctors to take steps to either diagnose or exclude a diagnosis of cervical cancer in the presence of a range of symptoms. While the relative costs and benefits of automated cytology remain a subject of considerable debate, and without government endorsement of these techniques, it seems highly unlikely that a court would expect a doctor to recommend such additional tests in an asymptomatic patient. The size of the problem needs to be kept in perspective. From Victorian Cytology Register figures correlating screening histories on women who die from cervical cancer, it has been estimated that at most, eight Victorian women dying from cervical cancer each year could identify laboratory error as a factor.¹¹ Adding $20 for ThinPrep and $30 for PapNet to each of the 600 000 smears done in Victoria each year would add $30 million to the laboratory costs of cervical screening.¹¹ Nationally this figure would amount to approximately $70 million, and the health care community must ask whether this is the best use of available resources. With some innovative consideration of alternatives, this money could be applied to other areas with likely better outcomes. For example, one alternative may be to pay GPs a Medicare rebate for taking a Pap smear on a previously unscreened older woman. Offering a $20 rebate for every 50-70-year-old unscreened woman in Australia would cost around five million dollars per annum, and could be expected to prevent a substantial number of cervical cancers in this group most "at risk". The case for a massive investment of public funds into these technologies appears small, and Australian taxpayers should not be expected to bear the full cost of their development when other health systems have failed to endorse them. While providing advice and information to patients is part of the standard of care, health practitioners are not agents for commercial enterprises. The right of individual women to spend their own money on items of their choice is to be respected, but their choice should be free and informed. The health gain from such a purchase needs to be clear to women and represented accurately: the available information on this technology suggests that the incremental benefit of automated cytology over conventional cytology is really quite small, and is unlikely to be substantially better than having a repeat Pap smear two years later.¹² The current marketing strategy runs the risk of reducing confidence in the existing system and of introducing apparent inequity into cervical screening -- a particular problem when screening rates are lowest and cervical cancer incidence is highest in women of low socioeconomic status.¹³ Finally, in the current medicolegal environment, suggesting that there may be legal implications for doctors who do not recommend these automated tests to their patients is unwelcome and unusually coercive. Gerard V Wain Director, NSW Cervical Screening Program, Westmead Hospital, NSW Mitchell HS, Giles GG. Cancer diagnosis after a report of negative cervical cytology. Med J Aust 1996; 164: 270-273. Koss L. The Papanicolou test for cervical cancer detection: a triumph and a tragedy. JAMA 1989; 251: 737-743. Cervical Screening in NSW: situation analysis, February 1997. Sydney: NSW Cervical Screening Program, 1997. Statement on technical devices for innovation in cervical cytology screening [editorial]. Am J Clin Pathol 1996; 106: 441. Farnsworth A, Chambers FM, Goldschmidt CS. Evaluation of the PAPNET system in a general pathology service. Med J Aust 1996; 165: 429-431. Roberts JM, Gurley AM, Thurloe JK, et al. Evaluation of the ThinPrep Pap test as an adjunct to the conventional Pap smear. Med J Aust 1997; 167: 466-469. Hutchinson M. Assessing the costs and benefits of alternative rescreening strategies. Acta Cytologica 1996; 40: 4-7. Assessment of techniques for cervical cancer screening. Ottawa: Canadian Coordinating Office for Health Technology Assessment, 1997. Gurley AM, Roberts JM, Thurloe JK, et al. Increasing the accuracy of the Pap test [letter]. Med J Aust 1997; 167: 507. Saunders C. Pap smear wizardry: new technologies raise difficult questions. Aust Doctor 1997; 27 June: 28-32. Check W. Too early to solve Pap device puzzle. CAP Today June 1997;11: 6. Mitchell H, Medley G. Detection of laboratory false negative smears by the PapNet cytological screening system. Acta Cytologica 1997. In press. Smith D, Taylor R, Coates M. Socioeconomic differentials in cancer incidence and mortality in urban New South Wales, 1987-1991. Aust N Z J Public Health 1996; 20: 129-137.

The Medical Journal of Australia eMJA

©MJA 1997