Volume 166 Issue 1

Can "best practice" be achieved outside metropolitan centres? 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/>". About five million Australians (almost 30% of the population) live outside metropolitan centres, in regional towns or more remote areas. Providing good health care for these people, spread over such a large area, has always been a challenge. Recently, concern has focused on access to general practitioner services in rural and remote areas, but access to specialist care is equally important. Surgeons in rural areas, unlike their metropolitan counterparts, are responsible for a broad range of surgical care and do not have the opportunity to concentrate exclusively on a narrow field of practice, such as breast surgery. In recognition of this, the Royal Australasian College of Surgeons (RACS) has established a Division of Rural Surgery, which provides a forum for discussion of rural surgical services and continuing medical education for surgeons who live and work at a distance from major teaching institutions. the expertise of a local specialist who understands the circumstances and needs of patients living in remote areas is invaluable in planning treatment Treatment of breast cancer in Australia has been extensively discussed recently, and a recent report 1 has emphasised the need for multidisciplinary management in centres which are, of necessity, in urban areas. However, a recent South Australian study showed that almost 25% of breast cancers occur in women who live outside urban areas, 2 and there are about 1.5 million women at risk of breast cancer in country areas (calculated from Australian Bureau of Statistics data for women aged over 30 years). BreastScreen Australia, which operates in every State and Territory, plans to screen about a third of these (including 70% of women aged 50-69 years), so that many new, early breast cancers will eventually be detected in rural women. Nevertheless, there is little information about the treatment of breast cancer in rural areas. Tulloh and Goldsworthy's study of breast cancer management in a Victorian country town (in this issue of the Journal) provides valuable information for the debate on how best to deliver treatment to women in rural areas. There is abundant evidence that breast-conserving surgery achieves disease control with good cosmesis in many breast cancer patients. No studies have shown any difference in long term survival rates between women who have had breast-conserving surgery and those who have had a mastectomy for a similar type and stage of disease, and breast-conserving surgery is widely accepted as the preferred treatment for early-stage breast cancer. 3 However, there is substantial geographic variation in the use of breast conservation as standard treatment in both Australia and the United States, 4,5 and there is concern that time, travel and communication may prove obstacles to such treatment in rural areas. Tulloh and Goldsworthy show that this need not be so. Experience at metropolitan centres in Australia shows that more than 50% of primary breast cancers are now treated conservatively; Tulloh's index of 68% is higher. Concern has also been expressed that limited experience in surgical management of breast cancer may be an impediment to "best practice". As recently as 1986, 25% of surgeons who treated breast cancer in Victoria treated fewer than five cases a year. 6 While most urban surgeons now agree that breast cancer should be treated in a multidisciplinary setting by a specialist who treats many breast cancers, almost 50% of rural surgeons believe this is not necessary (National Health and Medical Research Council National Breast Cancer Centre and RACS Joint Study, personal communication from Dr S Redman, Director, National Breast Cancer Centre, Sydney, NSW). This may be an issue for training of surgeons: those who have had appropriate experience during their higher surgical training are likely to retain competence, despite a relatively small workload. Training in breast surgery has been greatly improved by the development of specialist surgical breast units in metropolitan teaching hospitals. In addition, the establishment of reaccreditation and continuing education programs within the RACS ensures that every surgeon who practises breast surgery has the opportunity to be informed of new treatment developments. Although the rate of surgical complications reported by Tulloh was higher than would be expected for a specialist surgeon (those who treat more than 100 breast cancers a year would rarely see skin-flap necrosis after a mastectomy and would expect a haemotoma rate of 2%-3% after open biopsy), such rates are acceptable in a general surgical practice. The benefits of multidisciplinary management of breast cancer have been debated for some time. Multidisciplinary units in metropolitan centres are justified by their logistic advantages, by the opportunity to establish high standards with consistent peer review and by the small, but demonstrable, improvement in outcome. 7,8 The National Health and Medical Research Council clinical guidelines 9 espouse the principle of multidisciplinary care, but also point out that this can be achieved outside integrated treatment centres, by consultation between appropriate specialists. Tulloh and Goldsworthy have demonstrated that this can be done efficiently. Cost and logistics prevent the establishment of rural radiation treatment centres, but many medical and radiation oncologists visit regional centres for consultations and treatment planning. Electronic communication between rural and metropolitan centres can also facilitate multidisciplinary management. In this context, the expertise of a local specialist who understands the circumstances and needs of patients living in remote areas is invaluable in planning treatment. Australians who live in cities sometimes believe that those who live in the country accept that access to some components of health care is restricted. The determination of rural specialists and nurses like Tulloh and Goldsworthy to demonstrate that high standards of care, comparable to those in urban Australia, can be achieved in rural centres is an example to all health professionals. In addition, their demonstration that breast cancer can be managed effectively in rural centres serves as a model for other cancers and shows that, with effective communication, there is no insuperable barrier to high standards of health care in rural Australia. Colin M Furnival Senior Visiting Specialist, Surgical Breast Unit Royal Brisbane Hospital, QLD. House of Representatives Standing Committee on Community Affairs. Report on the management and treatment of breast cancer in Australia. Canberra: AGPS, 1995. South Australian Cancer Registry. Epidemiology of cancer in South Australia. Adelaide: The Cancer Registry, 1996: 211-236. Harries SA, Lawrence RN, et al. A survey of the management of breast cancer in England and Wales. Ann R Coll Surg Engl 1996; 78: 197-202. Byrne MJ, Jamrozik K, Parsons RW, et al. Breast cancer in Western Australia in 1989. II. Diagnosis and primary management. Aust N Z J Surg 1993; 63: 624-629. Farrow DC, Hunt WC, Samet JM. Geographic variation in the treatment of localised breast cancer. N Engl J Med 1992; 326: 1097-1101. Hill DJ, Giles GG, Russel IS, et al. Management of primary, operable breast cancer in Victoria. Med J Aust 1990; 152: 67-72. Sainsbury R, Haward B, Rider L, et al. Influence of clinician workload and patterns of treatment on surviving from breast cancer. Lancet 1995; 345: 1265-1270. Gillis CR, Hole DJ. Survival outcome of care by specialist surgeons in breast cancer: a study of 3786 patients in the west of Scotland. BMJ 1996; 312: 145-148. National Health and Medical Research Council. The management of early breast cancer. Clinical practice guidelines. Canberra: NHMRC/AGPS, Oct 1995.

Colin M Furnival

Audit of a general surgical practice in the rural Victorian town of Echuca identified 28 new patients with breast cancer between September 1992 and August 1995 (10% of those with breast conditions). The rural setting was no impediment to breast conservation (achieved in 68% of the 25 who had surgery) or to a multidisciplinary approach (management was planned in conjunction with an oncologist and/or specialist breast surgeon for 26 of the 28 patients). 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". Introduction - Setting - Patients - Surgical treatment - Specialist liaison - Other treatment - Complications - Long-term outcome - Discussion - Acknowledgements - References - Authors' details Introduction Carcinoma of the breast is a major public health issue at present. It has been the subject of two Government inquiries, 1,2 the more recent of which suggested that management of breast cancer across Australia was fragmented, uncoordinated and, at least in some areas, out of date. 1 The National Breast Cancer Centre 3 was established recently, with the mission of ensuring the highest standard of care for all Australian patients with breast cancer. The National Health and Medical Research Council has published guidelines to assist both patients 4 and doctors 5 in achieving this aim. These publications indicate that breast cancer management is highly specialised and imply that it is best done in major metropolitan centres where multidisciplinary expertise is available on-site. However, the need for specialised centres for breast cancer management is questionable. 6 Submissions from rural women to the Government inquiriesdescribed the increased burden imposed on them and their families in travelling for treatment and their consequent lack of treatment choices. 3 Many country people prefer local treatment and expect that their local hospital will provide the highest standard of care. However, to our knowledge there have been no publications on how breast cancer is managed in Australian rural centres. To provide this information, we conducted a three-year survey of breast cancer management in the practice of a general surgeon (B R T) in the rural Victorian town of Echuca. Setting Echuca is in northern Victoria, about 200 km from the capital, Melbourne. It has a 64-bed general hospital, which serves the town's population of about 15 000 and another 20 000 people in surrounding farming districts. During the survey period, formal links were in place with the oncology unit at St Vincent's Hospital, Melbourne, whose specialists regularly visited the regional base hospital at Shepparton, about 80 km from Echuca. More recently, the Bendigo Hospital, about 100 km away, has appointed a full-time medical oncologist, who now consults monthly at Echuca. These links have been important for interdisciplinary communication, as well as facilitating patient referral for radiotherapy (available in Melbourne) when required. Histology and cytology services are provided at Shepparton. A daily courier service transports specimens to the laboratory and reports are faxed within 48 hours. Hormone receptor studies on paraffin-embedded material with immunoperoxidase stains are available on request. Patients The study involved review of the medical records of all patients who attended the Echuca Consulting Suites under the care of a general surgeon (B R T) between 1 September 1992 and 31 August 1995. This is the surgeon's only practice location. Both private and uninsured patients were included, as the local hospital has no outpatient clinic. Surgical details were obtained from the surgeon's log book, and details of chemotherapeutic regimens and inpatient management from hospital medical records when required. Between September 1992 and August 1995, there were 1992 new patients (excluding emergencies and in-hospital consultations); 275 (14%) were seen for breast conditions. All of these were referred directly by their general practitioners. Surgery was offered for suspected malignancy and for a palpable breast lump that was either atypical on cytological examination, or was causing undue anxiety to the patient. Eighty-two patients (30% of those with a breast condition) had surgery; 57 had benign disease and 25 had cancer. Of those with cancer, 22 were diagnosed by fine-needle aspiration biopsy, 24 by mammography, and two by open biopsy (some patients had more than one form of test). Three patients with advanced breast cancer did not have surgery, giving a total of 28 with breast cancer (10% of the 275 with a breast condition). To our knowledge, none of the remaining 190 have since developed breast cancer. Nine of the 82 patients undergoing surgery had wire-localised excisions of lesions detected by mammography; four had carcinoma. Surgical treatment Treatment of the 25 patients who underwent surgery for cancer is summarised in the Box. Seventeen patients (68%) were suitable for breast conservation, having relatively small, peripheral tumours in breasts of moderate size. However, eight (32%) required mastectomy: two of these had large central tumours for which local excision would have been disfiguring; two had extensive in-situ components for which local treatment would have been inadequate; two required palliative mastectomy for control of locally advanced primary tumours; one presented with axillary lymphadenopathy alone; and one had pure ductal carcinoma-in-situ. The latter and the two having palliative surgery had simple mastectomies. The other five had modified radical mastectomies, with preservation of the pectoralis minor muscle. Clearance of axillary lymph nodes was undertaken for 15 of the 17 patients who had wide local excision of their primary breast cancers (for 12 this was done in Echuca and for three after they chose to go to Melbourne for further management). Two patients (aged 73 and 76, with tumours of 11 mm and 13 mm diameter, respectively) did not undergo axillary clearance, as it was considered that a positive lymph node yield would not alter their treatment regimen. They were given adjuvant tamoxifen alone. Altogether, 17 axillary clearances were done in Echuca (12 in association with a wide local excision and five with a modified radical mastectomy). The mean lymph node yield for those operated on in Echuca was 13 nodes (range, 4-29). Yields from the three patients who had axillary clearances by breast specialist surgeons in Melbourne were six, eight and 12 nodes, respectively. Specialist liaison Sixteen of the 28 new patients presenting with breast cancer were seen by an oncologist and/or specialist breast surgeon after diagnosis, sometimes before definitive surgery, to formulate a management plan (e.g., to discuss the role of axillary clearance or mastectomy). For a further ten, management plans were devised after telephone discussions with the other specialists. Specialist liaison was not obtained for two patients, as the surgeon was confident that the management plan was appropriate (and as one refused adjuvant therapy). Other treatment Each of the mastectomy patients had counselling and advice on prosthetics from the hospital's oncology nurse (M E G), who is also the prosthetics officer and stomal therapist. Each was invited to discuss referral for breast reconstruction, but so far only one has proceeded with this; she had a delayed transverse rectus abdominis myocutaneous (TRAM) flap reconstruction in Melbourne, with satisfactory results. Adjuvant radiotherapy was given to only nine of the 17 patients who underwent breast-conserving surgery. Seven of the other eight had small primary tumours (up to 12 mm diameter); the decision to omit radiotherapy was made for five of these by the oncologists consulted at the time or, for two, by the patients themselves, who chose not to travel 200 km to Melbourne for treatment. Another patient had undergone chest-wall irradiation in the past and was advised against having more. Twelve patients were referred for chemotherapy; one refused, despite exhaustive discussions with her general practitioner, surgeon and the regional oncologist, and 11 were seen by regional oncologists for planning of the chemotherapeutic regimen. At first, chemotherapy was given by the regional oncologists at Shepparton, but after 1993 it was given by the local oncology nurse (M E G) at Echuca under the supervision of the patient's general practitioner, with indirect supervision by the oncologist in charge. Two patients participated in national chemotherapy trials through the Bendigo Hospital Oncology Department; another patient was invited to participate but declined to do so. Adjuvant hormonal therapy was widely used. All post-menopausal patients with oestrogen receptor-positive tumours were prescribed tamoxifen. The decision to use this in patients with oestrogen receptor-negative tumours was made in conjunction with the regional oncologists. Complications Surgical complications were uncommon. One of the eight mastectomy patients developed superficial skin-flap necrosis, which healed satisfactorily without further surgery. Five of the 57 who underwent lumpectomy for benign disease developed a significant breast haematoma, defined as either a tense blood clot in the biopsy cavity or extensive bruising over the chest wall. Six of the 17 who had axillary clearance of lymph nodes developed seromas in the axilla, which required postoperative needle aspiration. One seroma became infected after aspiration, requiring intravenous antibiotics. Cosmetic results of surgery were acceptable to the patients and surgeon using subjective criteria (e.g., neat scar, no patient complaints). Careful consideration was always given to placement of incisions, design of mastectomy flaps, and use of drainage. 7 Fine absorbable subcuticular closure was used routinely. Long-term outcome Four of the 25 patients who underwent surgery for breast cancer have since died. All had advanced disease. Another patient, who was diagnosed with stage II disease and underwent local excision and chemotherapy, has developed metastatic disease. Discussion Our results show that a rural setting is no impediment to the use of breast-conserving surgery or a multidisciplinary approach in breast cancer management. Management of 26 of the 28 patients with breast cancer was planned in conjunction with an oncologist and/or specialist breast surgeon. Breast conservation was achieved in 68% of the 25 patients who underwent surgery. The frequencies of breast complaints (14% of new consultations) and of breast cancer (10% of breast complaints) in our practice were comparable with those found in two general surgical outpatient clinics in the United Kingdom. 8,9 However, the percentage of breast complaints that came to surgery in Echuca (30%) fell midway between the percentages found at Reading and Newbury in 1993 (16%) 8 and in London in 1982 (45%). 9 These differences could reflect changes in surgical policy with time, as well as differences in local criteria for biopsy. The ratio of benign to malignant lumps excised in Echuca was only 2:1, also midway between the figures for Reading and Newbury (1:1) 8 and for London (3:1), 9 probably for similar reasons. Modes of presentation and distribution across clinical stages in our series were similar to those found in a study of 856 patients with breast cancer from the Victorian Cancer Registry in 1990. 10 This study found that multidisciplinary management and use of breast conserving surgery was more likely for surgeons who treated more than 20 cases a year. In the Echuca series, with fewer than 10 cases a year, multi- disciplinary involvement and breast conservation rates were higher still. These differences reflect the different eras from which these reports have been derived, as breast cancer management strategies changed rapidly in the early 1990s, and make comparision of management strategies difficult to interpret. Are nine breast cancers a year enough for a centre to maintain competence in managing the disease? About 90 new patients with breast conditions (benign and malignant) were seen each year. Assuming each was seen two to four times in the year, there were 250-300 breast examinations per year, allowing expertise in examination to be maintained. Operative technique and complication rates were comparable to peer standards. 11 Management of breast cancer locally has numerous advantages for the country patient, not least of which is convenience. In fact, the travel involved (200 km) led one patient to refuse radiotherapy. On the other hand, local management has few, if any, disadvantages. Surgery for breast cancer is straightforward; more important are the management decisions. While "best practice" may require a multidisciplinary approach, we showed that this can easily be obtained by consultation with city colleagues on a case-by-case basis. Adjuvant therapy and follow-up can usually also be provided locally, provided that communication is maintained with the relevant specialists. Thus, it is important for rural doctors and nurses to keep up to date with trends in breast cancer management, but equally important that city-based centres of excellence encourage rural colleagues to participate in patient care. No patients in this series had their cancers detected by the regional mammographic screening program. Possibly, some patients from the Echuca region were identified in this way, but were referred to surgeons in Melbourne, as has occurred in other rural locations, leading to criticism. 12 However, as the standard of care provided in rural hospitals becomes acknowledged and patients' wishes about treatment location are taken into account, definitive management at local centres should become a policy for screening programs. Each of the potential inadequacies of isolated or small-volume practice (as identified in the House of Representatives Report on the management and treatment of breast cancer in Australia) 1 is addressed at Echuca, as is likely at many other country centres. Although relatively few new breast cancers were detected and treated each year, most were managed in consultation with other specialists. Operative technique and complication rates were comparable to peer standards; cosmetic results were considered satisfactory; patients had access locally to literature about breast cancer and treatment options, supportive counselling, a prosthetics service and chemotherapy and were offered referral for breast reconstruction after mastectomy; and several patients participated in trials of chemotherapy regimens. However, quality assurance was difficult; while surgical morbidity was monitored by the surgeon, formal review of adjuvant therapies' morbidity was lacking and round-table case discussions and histopathological or radiological review sessions were difficult to organise. It remains the responsibility of individual rural practitioners to seek such interdisciplinary contact as part of their continuing medical education. We conclude that a multidisciplinary service of high standard can be provided to patients with breast disease irrespective of their rural location, provided that they are prepared to travel for adjuvant therapy and for second opinions, if warranted. Acknowledgements The authors are indebted to Tamra Wines, Tracey McNair and Lisa Humphrys for their help in retrieving the data. References House of Representatives Standing Committee on Community Affairs. Report on the management and treatment of breast cancer in Australia. Canberra: AGPS, 1995. Senate Standing Committee on Community Affairs. Breast cancer screening and treatment in Australia. Canberra: Senate Printing Unit, 1994. Redman S, Kearsley JH. The National Breast Cancer Centre. Med J Aust 1995; 163: 432-433. National Health and Medical Research Council. Early breast cancer. A consumer's guide. Canberra: NHMRC/AGPS, Oct 1995. National Health and Medical Research Council. The management of early breast cancer. Clinical practice guidelines. Canberra: NHMRC/AGPS, Oct 1995. Ingram D. Who manages breast cancer? Aust N Z J Surg 1996; 66: 133. Dixon JM. Techniques of and indications for breast biopsy. Curr Pract Surg 1993; 5: 142-148. Dawson C, Lancashire MJ, Reece-Smith H and Faber RG. Breast disease and the general surgeon. Referral of patients with breast problems. Ann R Coll Surg Engl 1993; 75: 79-86. Cox PJ, Li MKW, Ellis H. Spectrum of breast disease in outpatient surgical practice. J R Soc Med 1982; 75: 857-859. Hill DJ, White VM, Giles GG, et al. Changes in the investigation and management of primary operable breast cancer in Victoria. Med J Aust 1994; 161: 110-122. Oertli D, Laffler U, Haberthuer F, et al. Perioperative and post-operative tranexamic acid reduces the local wound complication rate after surgery for breast cancer. Br J Surg 1994; 81: 856-859. Allsop JR. Changes in the investigation and management of primary operable breast cancer in Victoria [letter]. Med J Aust 1995; 162: 335. Authors' details Echuca Regional Health, Echuca, VIC. Bruce R Tulloh, MS, FRACS, General Surgeon; Marjorie E Goldsworthy, RN, Oncology Nurse and Breast Prosthetist, Division of Nursing. Reprints: Mr B R Tulloh, Corner of Francis and Leichhardt Streets, Echuca, VIC 3564.

Bruce R Tulloh · Marjorie E Goldsworthy

Exploring the analogy between computer viruses and biological viruses is a useful way to grasp the threat of "infection" to electronic medical records and data. 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". Introduction - Pathophysiology - Transmission - Symptoms and signs - Primary prevention - Secondary prevention - Detection - Treatment options - Conclusion - References Introduction Medicine is becoming increasingly dependent on information technology, 1-3 and clinicians should be aware of the risks posed by computer viruses. The effects of these viruses vary, ranging from irritating, but benign, messages on your screen like "Your computer is stoned" to permanent destruction of data. Not surprisingly, given their name, computer viruses are analogous to biological viruses, and analogies can be drawn between them in the prevention, detection and treatment of infection. The term "computer virus" was coined by Cohen in 1983, 4 although the notion of computer programs that could multiply was described by von Neumann in the 1940s. 5 It is difficult to establish exactly what the first computer virus was, as several existed before Cohen called them viruses. However, the first virus may have been a program called Creeper, 6 which spread over the ARPANET (the forerunner of the Internet) in 1970, announcing itself by "I'm the creeper . . . catch me if you can!". Cohen recognised that transmission of computer viruses was largely a "social" problem, 7 and has been proved correct. Pathophysiology Both computer and biological viruses are controlled by code -- computer code and genetic code, respectively. Genetic code evolves spontaneously to permit biological viruses to evade our immune system. Computer virus code is specifically crafted by programmers to exploit the weaknesses and idiosyncrasies of computer software, usually the operating system software. New biological viruses are uncommon, but ingenious computer programmers are always designing new viruses. Brand new viruses can easily evade a computer's defences, but fortunately are quite rare. However, while biological viruses are self-contained particles, able to exist in isolation under suitable physical conditions, computer viruses are code which must reside on some form of storage medium. The medium determines the longevity of the virus -- those on magnetic disks (floppy or hard) or tape persist until the medium is erased; those in computer memory survive only until the computer is turned off; and a virus on a CD-ROM (compact disc-read only memory) disk will last until the disk is destroyed . The term "infection" is used differently with biological and computer viruses. The mere existence of a biological virus does not qualify as infection. However, anything containing a copy of computer virus code is considered infected, be it the computer, its memory, floppy or hard disks, magnetic tapes, CD-ROMs or other storage media. Neither biological nor computer viruses can multiply without a functioning host. Computer viruses are able to replicate in a susceptible host computer, and virus copies are then stored in memory or on storage media. Computer viruses, like their biological counterparts, are species-specific. Computer species are defined by the program-running environment. For the vast majority of viruses, this is the operating system software (e.g., DOS, Windows NT, OS/2, Netware, Macintosh OS). So, for example, a Macintosh computer cannot catch a PC/DOS virus and vice versa. Newer "macro viruses" are specific to a particular application. For example, a "Microsoft Word" macro virus can spread from a PC to a Macintosh running Microsoft Word, but not to computers running other word-processing programs. There are many types and strains of computer virus, the number of which depend on a combination of the popularity of the computer and the ease with which its weaknesses are able to be exploited. There are over 1500 families of PC/DOS virus, fewer than 20 Macintosh virus families, and no known Unix viruses . Some computer viruses are very damaging, but these tend to disable their hosts before having much chance to replicate and be transmitted (like Ebola). Those which cause little damage tend to be more successful at propagation (like upper respiratory tract viruses). Some computer viruses create "carrier" states, where they may not reveal any overt evidence of infection, but are still infectious (like hepatitis B). Transmission Computer viruses are easily transmitted, as any means of transmitting computer code from one computer to another can pass on infection. Data files do not generally transmit viruses, because they do not contain executable code; the exception is the macro virus. Plain text e-mail messages are generally safe because they cannot transmit virus code. Binary files attached to e-mail messages or word-processor files containing macro viruses, however, may transmit virus code. Floppy disks are a common vector for computer viruses. Any means of transmitting computer code from one computer to another can transmit infection. Software downloaded from bulletin boards or the Internet could contain viruses, although systems operators try not to let viruses infect their systems. Computer networks can help viruses spread rapidly to all computers on the network. Symptoms and signs Each computer virus acts in a different way, so it is difficult to describe the behaviour of them all in general terms, except to say that they are capable of doing almost anything! Because they are computer programs, they do exhibit very deliberate behaviour. Detailed descriptions of known viruses are available in books, with antivirus programs, or on the Internet. The Internet is the best place to find out about viruses, because information on the latest viruses is available almost immediately they are documented in online virus bulletins or encyclopedias. While computer users often blame viruses for the odd behaviour of their computers, this is most commonly caused, not by infection, but by infestation -- with software "bugs" (faults in programs). The level of complexity in computer software these days is enormous, bugs occur frequently and bug-free software would be unaffordable for most people. (Mission-critical software, like that which controls jet aircraft, is meticulously designed, extensively tested and very expensive.) Primary prevention The best protection against computer virus infection is complete celibacy. For absolute protection against viruses, don't trade files with anyone; don't put your floppy disks in other people's computers; don't allow other people to put floppy disks in your computer; and don't connect your computer to any other computers. Computer viruses cannot appear spontaneously in a virus-free computer. However, while computer celibacy is safe, it is not for everyone. Just as it is impossible to simultaneously attempt natural pregnancy and avoid transmission of a sexually transmitted disease, computer celibacy is clearly impractical if you want someone else's file. "Safe data transfer" reduces the risk of viral transmission. Write-protecting your floppy disks (by sliding the write-protect tab on 3.5-inch diskettes as indicated by the arrow below) or other removable storage media will reduce the risk of transmission of a virus to your floppy. Because the write-protect function is managed by software, it may be disabled by a virus. Write-protecting a floppy disk does nothing to prevent transmitting a virus from this "donor" floppy. Secondary prevention Installing antivirus software is not unlike vaccination -- it can manually or automatically detect and prevent virus infection. Some antivirus programs try to detect "virus-like" activity and attempt to be a universal "vaccine". Unfortunately, it is impossible to reliably detect unknown (unidentified) computer viruses. Detection Early detection is important and, fortunately, antivirus software can also act as a screening test. Unlike the complexities of screening for human viruses, there is little effort involved in producing an antivirus program capable of detecting every currently known virus. Known computer viruses can be detected immediately, and all potential new contacts (i.e., floppy disks and downloaded files) should be screened with virus-detection software. Computer viruses can act in many different ways. When faced with the question "Does my computer have a virus?", the simple answer is to use up-to-date virus detection software as the arbiter -- if an antivirus program can't detect a virus, there probably isn't one. Clearly, it is essential that you always have the latest version of your antivirus software. Ingenious computer programmers are always designing new viruses. Luckily, those that can easily evade your antivirus defences are quite rare. There are good commercial shareware (try-before-you-buy) and freeware antivirus programs available, and those distributed by user groups or available on the Internet offer frequent updates. Treatment options A computer virus may be removed by either an antivirus program or by complete erasure of the infected storage medium. The importance of keeping adequate backup copies of data files cannot be emphasised strongly enough -- without them, your data may be lost forever. Appropriate treatment of a computer virus infection depends on the nature of the virus attack. "Benign" viruses spread without altering any user data, and hence can be completely removed from memory and disks by antivirus software without any loss of data. On the other hand, removal of malicious viruses can leave damaged program and data files or corrupted directories. The only way to recover the files is from backup copies made before the virus infection. In the worst cases, severe virus damage could necessitate complete reformatting of damaged disks (including hard disks), complete reinstallation of operating system software, and re-creation of data files from backup copies. It is vital to have "clean" original software distribution disks, with their write-protect tabs in the "protected" position. Unlike human viral infection, reinfection with computer viruses is common. Disinfecting a computer does not confer any "immunity". To prevent reinfection, it is wise to trace all computer contacts; that means any computer you have traded data with and all of the disks or other storage media you use with your computer. All contacts will need to be screened; those that test positive will need to be disinfected. Contact tracing can stop when all contacts are known to be virus free. Conclusion In general, it is unlikely your computer will catch a virus from original software distribution disks, especially from reputable companies. However, viruses have been spread on original software disks in the past, possibly by disgruntled employees, so you can never be too careful. Also, CD-ROMs are not immune from viruses, although the companies that produce them try hard to prevent viruses getting onto CD-ROMs in the first place. Lastly, and most importantly, we must recognise the value of our data and the importance of protecting it. To this end, practical guidelines are summarised above. While continued vigilance is essential as computer viruses often strike when you are least expecting them, they are only one small threat to electronic information Information security is a complete discipline in itself, 8 but that's another story . . . References Patterson R. Cyberspace surgery. Can Med Assoc J 1994; 151: 639-642. Cohen JJ. Educating physicians in cyberspace. Acad Med 1995; 70: 698. Coons T. Teleradiology: the practice of radiology enters cyberspace. Radiol Technol 1995; 67: 125-140. Fites PE, Johnston P, Kratz M, et al. The computer virus crisis. New York: Van Nostrand Reinhold. 1989. McAfee J, Haynes C. Computer viruses, worms, data diddlers, killer programs, and other threats to your system: what they are, how they work, and how to defend your PC, Mac or mainframe. New York: St Martin's Press, 1989. Lundell A. Virus -- the secret world of computer viruses that breed and destroy. Chicago: Contemporary Books, 1989. Cohen F. Short course on computer viruses. New York: Wiley, 1994. Caelli WJ, Longley D, Shain M, et al. Information security handbook. England: MacMillan, 1994.

Ian R Cheong · Michael R Kidd