Volume 172 Issue 7

Research Prevention of perinatal group B streptococcal disease: screening practice in public hospitals in Victoria Mary Connellan and Euan M Wallace MJA 2000; 172: 317-320 For editorial comment, see Oats Abstract - Introduction - Methods - Results - Discussion - Acknowledgements - References - Authors' details - - More articles on Infectious diseases and parasitology Abstract Objectives: To survey clinical protocols for prevention of early-onset group B streptococcal disease (EOGBSD) of the newborn in public maternity hospitals. Design: Postal questionnaire with telephone follow-up when required. Setting: All hospitals that undertook deliveries in public patients in the State of Victoria, November 1997 to January 1998. Results: The survey was sent to 84 hospitals: 71 responded and 64 met the criteria and provided usable data (76% response rate). These 64 represented 42 784 births (68% of births in Victoria in 1996). Most hospitals (62; 97%) undertook actions that would identify and treat pregnant women at risk of EOGBSD. 48 (75%) performed bacteriological screening for maternal GBS carriage, but only 20 of these had a unified protocol. Screening was mostly by low vaginal swab (15 hospitals) and before 30 weeks' gestation (12 hospitals). Low vaginal swab plus anal swab was used in only one hospital. Bacteriological screening was significantly more common in metropolitan hospitals than in rural hospitals (100% versus 67%; P = 0.007, Fisher's exact test). Targeting of prophylaxis by recognised risk factors was reported by 59 (92%) hospitals, 45 of which also undertook screening. There was considerable variation in the specific risk factors used. Conclusions: While there was clearly widespread awareness of EOGBSD in Victorian public hospitals, prevention programs varied considerably. The development of consensus practice guidelines might improve EOGBSD prevention, reducing morbidity, mortality and costs. Introduction Since the 1970s, group B streptococci (GBS) have been recognised as a major cause of neonatal systemic infection in the first week of life -- so called early-onset group B streptococcal disease (EOGBSD). The reported incidence of this condition varies between 1 and 4 per 1000 livebirths.1 Infants acquire the infection by vertical transmission from an asymptomatic mother during delivery.2 Clinical disease manifests at birth or within 24-48 hours as pneumonia, septicaemia or, less commonly, meningitis.1In Australia, the prevalence of GBS vaginal carriage has been estimated at 12%-15%,3-5 and about 1%-2% of infants born to women carrying GBS develop EOGBSD, with about 6% of cases being fatal.2,5-7The risks of EOGBSD and death are particularly high in preterm infants.2 However, antibiotic prophylaxis given to "at risk" women during labour has been shown to significantly reduce the incidence of EOGBSD, and is an important and worthwhile public health measure.2,7-9 While the value of prophylactic antibiotic intervention in at-risk women is now widely agreed, the best means of targeting these women is perhaps less clear. Comprehensive reviews of the available evidence have been published both in Australia1,5,10 and overseas.2,11-13 There are two broad approaches to targeting prophylaxis -- identification of GBS carriers by bacteriological screening or treating by clinical risk factors. No trials have compared the efficacy of the two approaches. It is therefore perhaps not surprising that, anecdotally, GBS intervention practices differ greatly between public hospitals across the State of Victoria. However, no objective data are available to assess the extent of these differences and the appropriateness of current practice. We surveyed all public maternity hospitals in Victoria to explore what GBS intervention programs were in place and, in particular, to assess whether practice was in line with currently available evidence. Methods The Victorian Perinatal Data Collection Unit, Melbourne, provided contact details of all maternity hospitals in Victoria and identified those that undertook deliveries in public patients. Between November 1997 and January 1998, a six-page survey form containing 18 questions was sent to the Delivery Suite Nursing Unit Manager, or equivalent, in each of these hospitals. The survey was multiple-choice format with some free-text fields. Non-respondents were sent a second copy of the survey form two months later and were telephoned if necessary. Data on deliveries in 1996 were supplied by the Victorian Perinatal Data Collection Unit. Statistical analyses were performed using Statview 4.1.14 Significance was taken as P < 0.05. Results Of the 84 hospitals surveyed, 71 responded and 64 met the criteria and provided usable data, giving a final response rate of 76% (three respondents delivered only private patients, one cared for postnatal women only, and three did not provide GBS screening information). The 64 hospitals that provided usable data accounted for 42 784 births in 1996 (68% of all births and 93% of all births to public patients in Victoria) and comprised 16 hospitals in metropolitan Melbourne and 48 rural hospitals. Of the 64 hospitals, 62 (97%) reported undertaking procedures that would identify and treat at least some women with a baby at risk of EOGBSD; 48 hospitals (75%) undertook routine antenatal screening for maternal GBS carriage, including 45 which also offered antibiotic prophylaxis on the basis of risk factors. Another 14 hospitals (22%) used the latter approach alone. Screening for GBS The hospitals which undertook routine antenatal bacteriological screening accounted for 97% of all deliveries in the 64 responding hospitals. They comprised all 16 metropolitan hospitals and 32 of the 48 rural hospitals, a significant difference in proportions between metropolitan and rural hospitals (P = 0.007, Fisher's exact test). Of the 48 hospitals that undertook screening, 20 had a unified hospital screening protocol, with the remainder using individual-doctor protocols. The 20 with a unified protocol comprised nine of the 16 metropolitan hospitals and 11 of the 32 rural hospitals (P = 0.22, Fisher's exact test). Characteristics of the screening protocols among these 20 hospitals are shown in Box 1. Most protocols (65%) were less than five years old, and 40% were less than two years old. The most common approach to bacteriological screening was to perform a low vaginal swab only (15 of 20 hospitals), or, less commonly, a high vaginal swab only (four hospitals). Only one hospital performed a low vaginal swab combined with an anal swab. All but one hospital screened only once in the pregnancy, either before 30 weeks' gestation (12 hospitals) or between 30 and 34 weeks' gestation (seven). The hospital that screened more than once did not specify gestations. All hospitals that performed bacteriological screening administered intrapartum antibiotics to all women who were GBS-positive. Screening of private patients Bacteriological screening was offered to private patients by some or all obstetricians at 37 of the 64 hospitals. This was a smaller proportion of hospitals than offered screening to public patients, although the difference did not reach significance (P = 0.06, Fisher's exact test). In only 18 of these 37 hospitals did all obstetricians offer screening to their private patients. In 13 hospitals, none offered screening to private patients, and in 14 the respondent did not know if it was offered. Risk-factor-targeted prophylaxis Targeting of prophylaxis by recognised clinical risk factors was reported by 59 of the 64 hospitals (92%). Criteria used are shown in Box 2. Antibiotics were reported to be given most commonly for clinical signs of intrapartum infection (51 hospitals) and pre-labour rupture of the membranes (43 hospitals), although the time from membrane rupture to starting antibiotic administration varied considerably. Only 10 hospitals administered antibiotics to women admitted in preterm labour below a specified gestation. All were metropolitan hospitals that also performed routine bacteriological screening. Only four hospitals (6%) administered antibiotics on the basis of all five recognised criteria. Two of these also undertook screening. With regard to the antibiotic used as chemoprophylaxis, 14 of the 20 hospitals with a unified protocol used penicillin, four amoxycillin and two ampicillin. The 14 hospitals using penicillin all had different treatment regimens. Of the hospitals that lacked a unified protocol, only four reported the antibiotic regimen used -- penicillin in three and amoxycillin in one. Discussion To our knowledge, this is the first survey of GBS screening practices in pregnancy to be reported in Australia. It reveals that prenatal screening and prophylaxis for GBS infection were widely practised in public hospitals in Victoria. However, the specific strategies varied considerably, and, while this variation is understandable (given the lack of robust comparative data for the various possible approaches2,10,12,15), it translated into a less than ideal approach in many centres. Current evidence suggests that the optimum approach to reduce EOGBSD is to offer intrapartum chemoprophylaxis, using penicillin (or erythromycin in women allergic to penicillin) to at-risk mother-infant pairs. These at-risk pairs are identified by bacteriological screening, involving a low vaginal and anal swab performed at 36-38 weeks' gestation and/or by clinical risk factors (Box 3).2,10 In our survey, most hospitals reported targeting prophylaxis through bacteriological screening. However, the varied approaches to this screening revealed that current practice may not be as effective as possible. In addition, while most hospitals undertook screening, only 20 had a unified protocol, while the remaining 28 reported that protocols differed between doctors. This, together with the variable practice for private patients, suggests that it may be useful to develop more uniform Australian guidelines. Indeed, that only two hospitals (3% of respondents) reported a current protocol that would be expected to maximally prevent EOGBSD (targeting prophylaxis by bacteriological screening and by all risk factors shown in Box 3) suggests that the introduction of uniform practice guidelines would be worthwhile. The most common differences between the reported screening protocols and an approach expected to minimise EOGBSD were the maternal sites sampled and the timing of screening. GBS carriage within individuals is not constant. Consequently, bacteriological swabs taken at 28 weeks' gestation have only a 50%-70% positive predictive value for carriage at delivery, while about 5%-10% of women who are GBS-positive at delivery are negative at 28 weeks.3,13,16 Therefore, the closer to delivery that bacteriological screening is undertaken, the greater its utility, both as sensitivity and specificity are increased,2,10,16 and as the costs of screening are saved for the 5% of pregnant women who deliver preterm (and should receive prophylaxis irrespective of the screening result.2,11). Thus, bacteriological screening is probably best undertaken at 36-38 weeks' rather than at less than 30 weeks' gestation, the most popular time in our survey. Nevertheless, despite these theoretical considerations, a significant reduction in the incidence of EOGBSD was recently reported by King George V Hospital, Sydney, where screening is performed at 28 weeks' gestation.17 This result emphasises that bacteriological screening at 28 weeks' gestation is preferable to no screening at all. Detection of GBS is increased by 5%-25% if an anal swab is collected in addition to a vaginal swab.16,18,19 Only one hospital in our survey reported collecting both swabs; most took only a low vaginal swab. It has been suggested that Australian women would find collection of anal swabs unacceptable,20 but no objective evidence has been presented for this. Furthermore, most United States centres surveyed by the Centers for Disease Control took anal swabs,21 suggesting that the practice may be more acceptable than is assumed. It would certainly be worthwhile asking Australian women, and reappraising the method of screening most appropriate for our population. Collection of high vaginal swabs, reported by four hospitals in our survey, is inappropriate for GBS screening. The most cost-effective approach to preventing EOGBSD is to offer antibiotic prophylaxis to women with identified risk factors, without bacteriological screening.15 This approach has been recommended by some Australian groups (Professor James King, Mater Perinatal Epidemiology Unit, Mater Misericordiae Mothers' Hospital, Brisbane, Qld, personal communication). Most hospitals surveyed (92%) offered prophylaxis on the basis of risk factors, but only four (6%) used all recognised risk factors appropriately (Box 3). As the relative risk of EOGBSD is significantly greater in preterm neonates than in babies born at term,7 any EOGBSD prevention program should ideally include intrapartum prophylaxis for any woman labouring before 37 weeks' gestation, irrespective of whether she was screened earlier in pregnancy or of the result of that screening.2,11 It was disappointing that only a minority of the hospitals surveyed had such a policy, particularly as infection per se is a major recognised cause of preterm labour. That no rural hospital had such a policy may reflect that these hospitals transfer such women for level 3 neonatal care, and that preparation for transfer focuses more on tocolytic therapy and corticosteroid prophylaxis. If so, then an educational campaign to encourage early antibiotic treatment instituted at the referring hospital might be worthwhile. We were also surprised that only a few hospitals offered antibiotic prophylaxis to women who had had a previous baby with EOGBSD. Neonatal EOGBSD is a devastating infection, and, in our experience, parents who have had an infected child usually seek interventions to prevent infection in a future delivery. That most clinicians and hospitals do not routinely offer prophylaxis in this situation suggests a lack of awareness that this history is an important risk factor. Similarly, the variable responses to the other accepted risk factors suggest that a significant proportion of health providers are either unaware of the epidemiology of EOGBSD or do not perceive EOGBSD as an important clinical problem. Our survey of Victorian public hospitals showed that, while most are clearly aware of EOGBSD, only a minority currently have a strategy that maximises prevention of EOGBSD and represents most cost-effective practice. However, very minor changes in practice would be expected to improve EOGBSD prevention and significantly reduce costs,15 for both screening and treatment. Accordingly, our data support the case for a comprehensive, statewide, or possibly national, education program, and for the development and introduction of uniform consensus practice guidelines. Acknowledgements The authors would like to thank the participating hospitals as well as Dr Jane Halliday and Ms Sofia Mercer, from the Victorian Perinatal Data Collection Unit. EMW was partly funded by a Charles and Sylvia Viertel Clinical Investigator Fellowship. Disclosure: We are not aware of any conflict of interest arising from performing or reporting this work. References Vigneswaran R, O'Loughlin JA, McDonald HM. Group B streptococcus and pregnancy. Aust N Z J Obstet Gynaecol 1995; 35: 117-119. Prevention of perinatal group B streptococcal disease: a public health perspective. Centers for Disease Control and Prevention. MMWR Morb Mortal Wkly Rep 1996; 45 (RR-7): 1-24. McDonald H, Vigneswaran R, O'Loughlin JA. Group B streptococcal colonization and preterm labour. Aust N Z J Obstet Gynaecol 1989; 29: 291-293. Australasian Study Group for Neonatal Infections. Early-onset group B streptococcal infections in Aboriginal and non-Aboriginal infants. Med J Aust 1995; 163: 302-306. Jefferey HE, McIntosh ED. Antepartum screening and non-selective intrapartum chemoprophylaxis for group B streptococcus. Aust N Z J Obstet Gynaecol 1994; 34: 14-19. Garland SM, Fleigner JR. Group B streptococcus (GBS) and neonatal infections: the case for intrapartum chemoprophylaxis. Aust N Z J Obstet Gynaecol 1991; 31: 119-122. Zangwill KM, Schuchat A, Wenger JD. Group G streptococcal disease in the United States, 1990: report from a multistate active surveillance system. Morb Mortal Wkly Rep CDC Surveill Summ 1992; 41: 25-32. Smaill F. Intrapartum antibiotics for Group B streptococcal colonisation (Cochrane review). The Cochrane Library, 1999: 3. Oxford: Update Software. Schrag SJ, Zywicki S, Farley MM, et al. Group B streptococcal disease in the era of intrapartum antibiotic prophylaxis. N Engl J Med 2000; 342: 15-20. Gilbert GL, Isaacs D, Burgess MA, et al. Prevention of neonatal group B streptococcal sepsis: is routine antenatal screening appropriate. Aust N Z J Obstet Gynaecol 1995; 35: 120-126. Schuchat A. Group B streptococcus. Lancet 1999; 353: 51-56. Rouse DJ, Goldenberg RL, Cliver SP, et al. Strategies for the prevention of early-onset neonatal group B streptococcal sepsis: a decision analysis. Obstet Gynecol 1994; 83: 483-494. Regan JA, Klebanoff MA, Nugent RP, et al, VIP Study Group. Colonization with group B streptococci in pregnancy and adverse outcome. Am J Obstet Gynecol 1996; 174: 1354-1360. Statview 4.1. Berkeley, CA: Abacus, 1994. Garland SM, Kelly N. Early-onset group B streptococcal sepsis: economics of various prevention strategies. Med J Aust 1995; 162: 413-417. Boyer KM, Gadzala CA, Kelly PD, et al. Selective intrapartum chemoprophylaxis of neonatal group B streptococcal early-onset disease. II. Predictive value of prenatal cultures. J Infect Dis 1983; 148: 802-809. Jefferey HE, Lahra MM. Eight-year outcome of universal screening and intrapartum antibiotics for maternal group B streptococcal carriers. Pediatrics 1998; 101: E2. Badri MS, Zawaneh S, Cruz AC, et al. Rectal colonisation with group B streptococcus: relation to vaginal colonisation of pregnant women. J Infect Dis 1977; 135: 308-312. Dillon HC, Gray E, Pass MA, Gray BM. Anorectal and vaginal carriage of group B streptococci during pregnancy. J Infect Dis 1982; 145: 794-799. Prevention of neonatal group B streptococcal sepsis: is routine antenatal screening appropriate? [editorial comment]. Aust N Z J Obstet Gynaecol 1995; 35: 120. Adoption of hospital policies for prevention of perinatal group B streptococcal disease -- United States, 1997. MMWR Morb Mortal Wkly Rep 1998; 47: 665-670. (Received 10 Nov 1999, accepted 14 Feb 2000) Authors' details Monash Medical Centre, Melbourne, VIC. Mary Connellan, RM, MPH, Midwife, Women's Health Program, Southern Healthcare Network; Euan M Wallace, MD, FRACOG, Senior Lecturer, Department of Obstetrics and Gynaecology, Monash University. Reprints: Dr E M Wallace, Department of Obstetrics and Gynaecology, Monash University, Monash Medical Centre, 246 Clayton Road, Clayton, VIC 3168. euan.wallaceATmed.monash.edu.au Make a comment 1: Characteristics of screening for group B streptococcus in 20 hospitals with a unified screening protocol Number of hospitalsYears since screening protocol introduced < 28 (40%) 2-55 (25%) > 5-91 (5%) > 92 (10%) Not known4 (20%) Form of screening Low vaginal swab only15 (75%) High vaginal swab only4 (20%) Low vaginal swab and anal swab1 (5%) Frequency per pregnancy Once only19 (95%) More than once1 (5%)Timing < 30 weeks' gestation12 (60%) 30-34 weeks' gestation7 (35%) Unknown1 (5%)**The hospital that screened more than once did not specify gestations. Back to text 2: Criteria for intrapartum administration of antibiotics in 64 hospitals in Victoria CriteriaNumber of hospitals usingClinical signs of intrapartum infection51 (80%)Pre-labour rupture of membranes43 (67%) < 6h4 (6%) 6-12h0 > 12-18h10 (16%) > 18-24h4 (6%) > 24h21 (33%) Doctor-dependent4 (6%)Previous GBS-affected baby30 (47%)GBS-positive vaginal swab in previous pregnancy22 (34%)Preterm labour10 (16%) Gestation < 37 weeks7 (11%) Gestation < 34 weeks2 (3%) Gestation < 32 weeks1 (2%)GBS=Group B streptococcus. Back to text 3: Key risk factors for early-onset neonatal group B streptococcal disease (EOGBSD) Preterm delivery (< 37 weeks' gestation) Prolonged rupture of membranes (> 18h) Previous infant with EOGBSD GBS bacteriuria during pregnancy Intrapartum maternal pyrexia Back to text

Mary Connellan · Euan M Wallace

Viewpoint Medicine for the millennium: the challenge of postmodernism Jonathan J Chan and Julienne E Chan MJA 2000; 172: 332-334 Abstract - Postmodernism - Evidence-based medicine - Clinical pathways and diagnosis-related groups - The future - Acknowledgements - References - Authors' details - - More articles on General medicine Abstract As the new millennium dawns, Australian society is becoming more postmodern, whereas the medical system remains increasingly modernist in its outlook. In this article, we discuss the emerging prevalence of postmodernism and examine current medical education and practice strategies, such as evidence-based medicine, from a postmodern perspective. We argue that if medicine does not respond to the ideas of postmodernism, which challenges the concepts of truth and our ability to be objective, it may become increasingly irrelevant to the needs of a changing society. Examining the state of medicine in a postmodern world is important to a profession increasingly reliant on science and technology within a society increasingly distrustful of such a modernist approach. Does medicine run the risk of becoming outmoded in the face of the postmodern expectations of its patients? Postmodernism This amorphous thing [postmodernism] remains ghostly -- and for some, ghastly -- for the simple reason that the debate around the postmodern has never properly been engaged. Thomas Docherty1 Asking the question "what is postmodernism?" is a (post)modern-day equivalent of trying to capture Proteus, "for the concept is not merely contested, it is also internally conflicted and contradictory . . . Postmodernism is not something we can settle once and for all and then use with a clear conscience".2 The term "postmodernism" itself inevitably leads us back to modernism -- whether it be to replace it, reject it, re-evaluate it or revitalise it. Modernism can, in short, be characterised by belief in the existence of truth, objectivity, determinacy, causality and impartial observation.3 It has been described as "a search for an underlying and unifying truth and certainty, a search for a definitive discourse that makes the world and self coherent, meaningful and masterable".4 Modernism thus seeks to capture, define, understand and control knowledge. To return to postmodernism, definitions abound. It has been variously described as an epoch or historical period;1 a theoretical and representational mood, a cultural epoch and an aesthetic practice;5 a sensibility;6 a consciousness, the cultural logic of late capitalism and the crystallisation of previously independent developments;2 a number of related tendencies, values, procedures and attitudes;7 and the general condition of contemporary Western civilisation.8 These definitions are equally valid, inclusive, exclusive, overlapping, complementary and contradictory. For example, in contrast to descriptions of postmodernism as a historical period, or a consciousness, Bauman argues that the notion of historical succession is an illusion, and that the postmodern era is a philosophical and sociological re-evaluation of modernity.9 Jameson goes further, to suggest the disappearance of a sense of history in favour of perpetual change in a perpetual present.10 While these definitions encompass the historical, the aesthetic, the philosophical, the sociocultural, and the politico-economic, Lyotard has described postmodernism as "also, or first of all, a question of expressions of thought".11 From this perspective, one view of postmodernism is that it is the third stage in the evolution of Western conceptions of knowledge, society and culture. The premodern, or classical, era was based on the spiritual and the mythological, and can be summarised by Anselm's credo "I believe so that I might understand".12 The gods (and later God), not humanity, were the centre of the universe. The modern age of the Enlightenment saw a radical change in perspective as humanity, empowered by science and reason, took centre stage. Descartes' axiom "I think, therefore I am" resounded12 as humankind grabbed the keys and set about unlocking the doors, discovering what lay behind them, and determining the boundaries of knowledge. As humanity's glorious hopes for self-determined science, progress and freedom waned, disappointment and disillusion set in. At the turn of the century, the prospect of the new age of technology heralded hopes for a better world, free of disease and social inequality. Yet, at the dawning of a new millennium, the fact that these promises have not been fulfilled has led to increasing doubt about the ability of science to heal and liberate. Although science has generally improved human health and comfort, scientific advances, such as the prolongation of human life, have resulted in a plethora of other problems which medicine and science have difficulty addressing. The widespread use of unconventional therapies in chronic illnesses such as cancer13 and arthritis14 shows that patients are seeking treatment which conventional scientific medicine cannot provide. Bauman has suggested that Wittgenstein's description of understanding as "knowing how to go on"9 encapsulates this current era. The evolution from believing to knowing the facts leaves us at a point of knowing from experience, with the credo "I experience, therefore I try to make sense". In line with a move away from the overarching themes and theories of the Enlightenment, Lyotard has suggested that postmodernism can be simplified to a disbelief in métarécits, or philosophical metanarratives, such as "Science"and "Truth", in favour of the petit récit, the small narrative based on lived lives, the diverse, the complex and the unique.8 Such an approach acknowledges individuality, complexity and the subjectivity of personal experience. The postmodernist paradigm cannot accept that all things may be understood and mastered through science. The validity of intuition and experience is considered equal to that of traditional methods of observation, induction and experimentation. If society believes that the rational, objective truths and certainties of science and medicine are not as true and not as certain as they once may have seemed, where does that leave a practice of medicine which continues to base itself on a modernist approach? Medicine is resolutely progressing down a path of innate modernity. We discuss two aspects of this trend: evidence-based medicine and clinical pathways. Evidence-based medicine is modernist medicine He's the best physician that knows the worthlessness of most medicines. Benjamin Franklin15 Much has already been written about the benefits and caveats of evidence-based medicine (EBM). EBM considers patient management based on available data and medical literature as conceptually vital and important for best care. Since its inception in the early 1990s, EBM has had widespread impact on the teaching and practice of medicine. A postmodernist would regard EBM and the value it places on what is considered to be "current knowledge" as a modernist concept. A postmodernist would question whether current science and technology have the ability to give us the "evidence" vital to the practice of EBM. Thus, EBM is very likely flawed given that "the evidence is based only on our current value systems, which can dramatically alter with new advances in our understanding of nature".16 Proponents of EBM would argue that the constant search for current data would ensure that the practice of medicine is kept abreast of whatever new trends may occur. However, another disturbing consequence of EBM is not only the quest for the right sort of data, but also the essence of the data itself. EBM journals are edited by combinations of physicians, epidemiologists and other experts, who determine the importance of the research. Currently, 98% of articles reviewed are rejected.17 Already, there is a worrying trend that only well-funded, large, multicentre trials are published in first-rank, high-impact-factor journals. Many recent advances in clinical care have been determined from pharmaceutical trials. How do paradigm shifts occur when the motivation for research is biased, not towards "best evidence", but rather to that which would guarantee high-profile publication or sufficient pharmaceutical sales? The recent furore over a high-profile researcher who was not allowed to publish her findings because they contradicted claims about the therapeutic efficacy of the products of her pharmaceutical funding body is an example of how research is increasingly driven by profit.18 In addition, examination of medical literature shows a paucity of articles which report negative findings or use qualitative research methods. Surely the determination of "best evidence" requires consideration of such data? The question is therefore, not "what is 'best evidence'", but "how is 'best evidence' determined" and "is it really the 'best' evidence"? Most physicians would argue that the practice of medicine is an art -- an ill-defined combination of experience and judicious use of knowledge. EBM teaching emphasises "knowledge" -- learning the "facts" and knowing the "literature". Sackett et al recognised this in their own exhortations that EBM is not "cookbook medicine", and that the "external clinical evidence can inform but not replace individual clinical expertise and it is this expertise that decides whether the external evidence applies to the individual patient at all".19 However, in teaching EBM to medical students, there is a danger of "dumbing-down" medicine to the lowest common denominator of understanding facts and applying treatment algorithms without applying Sackett's caveats. The emphasis placed on acquiring medical knowledge may produce practitioners who have no understanding of the uniqueness of each patient. Clinical pathways and diagnosis-related groups are modernist ideals I am truly horrified by the modern man. Such absence of feeling, such narrowness of outlook, such lack of passion and information, such feebleness of thought. Alexander Herzen20 One of the major arguments against modernism and its advances is the dehumanisation of society. From a postmodernist view, the individual is now a faceless number in the databank of society. Technological advancement and efficiency "ha[ve] left people feeling disconnected with one another".12 Campion, in an editorial on "Unconventional medicine",21 posits that, "though Americans want all that modern medicine can deliver, they also fear it. They may resent the way that visits to physicians quickly lead to pills, tests, and technology . . . [they] also may seek out unconventional healers because they think their problems will be taken more seriously". In an effort to rationalise the growing health budget, Australian health providers are now determining costs through funding by classification of diseases through diagnosis-related groups (DRGs) and clinical pathways. The benefits are obvious: greater efficiency in treatment has meant reduction of hospital waiting lists, reduced hospital stays and reduction of costs. The downside is the dehumanisation feared by the postmodernist. Proponents of the system argue that the benefits of more people being treated outweigh the apparent loss of identity. Yet, the outcome of this method of medicine is far more sinister than it seems. The loss of identity and the classification of admissions as DRGs have resulted in a health system which encourages medical practitioners to focus only on disease and to fail to understand the individuality and uniqueness of each patient. The fact is that DRGs and clinical pathways are preparing a future generation of medical practitioners who will be very specialised in treating patients according to such pathways, but little prepared for significant deviations from them. Clinicians are becoming very adept at procedures and skills determined by diseases and not by the individual patient's signs and symptoms. The future There is an ominous cloud in the distance though at present it be no bigger than a man's hand. Arthur Stanley Eddington22 Medicine is continually changing. It came into existence with the Enlightenment and gained scientific maturity in the modernist age. Yet, the current foundation of medical knowledge (EBM) and its essence of practice (DRGs, clinical pathways) are significant constraints which will inhibit its ability to change with the times. In effect, medicine is becoming a modernist phenomenon which can neither progress nor provide the necessary service to a society which is increasingly postmodernist. In the past, there were fewer alternatives to medical practice. Nowadays, the needs of society are met by allied health professionals, naturopaths and other, similar therapists. The role of the medical practitioner is already changing. Doctors are now "healthcare providers" who administer "health services". Patients are now "clients". It is likely that the medicine we know will become just one part of a holistic health service which includes other practitioners currently regarded as "alternative". Producing medical practitioners who know only clinical pathways and DRGs further widens the gap between the modernist model of dehumanised science (the grand narrative) and the postmodernist model of unique, lived experience (the small narrative). Unless the practice of medicine becomes more focused on the unique individual, with understanding of the limitations of the modern science of medicine, our role runs the risk of becoming less relevant to people today. Acknowledgements Jonathan Chan was supported by the Janssen-Cilag Dermatology Research Fellowship of the Australasian College of Dermatologists and the Amy and Athelstan Saw Postgraduate Research Fellowship of the University of Western Australia. Julienne Chan was supported by an Australian Postgraduate Research Award. References Docherty T. Postmodernism: an introduction. In: Docherty T, editor. Postmodernism: a reader. New York: Harvester Wheatsheaf, 1993: 1-31. Jameson F. Postmodernism, or the cultural logic of late capitalism. London: Verso, 1991. Rosenau P. Post-modernism and the social sciences: Insights, inroads and intrusions. Princeton, New Jersey: Princeton University Press, 1992. Usher R, Bryant I, Johnston R. Adult education and the postmodern challenge: learning beyond the limits. London: Routledge, 1997. Waugh P, editor. Postmodernism: a reader. London: Edward Arnold, 1992. Laclau E. Politics and the limits of modernity. In: Docherty T, editor. Postmodernism: a reader. New York: Harvester Wheatsheaf, 1993: 329-343. Hassan I. Toward a concept of postmodernism. In: Docherty T, editor. Postmodernism: a reader. New York: Harvester Wheatsheaf, 1993: 146-156. Lyotard J-F. The postmodern condition: a report on knowledge. Bennington G, Massumi B, translators. Manchester: Manchester University Press, 1984. Bauman Z. Philosophical affinities of postmodern sociology. Sociological Rev 1990; 38: 411-444. Jameson F. The cultural turn: selected writing on the postmodern, 1983-1993. London: Verso, 1998. Lyotard J-F. Note on the meaning of "post-". In: Docherty T, editor. Postmodernism: a reader. New York: Harvester Wheatsheaf, 1993: 47-50. Utley D. Communicating truth in the post-modern world. Luke's Journal 1998; 3: 8-9. Group CO. New Zealand cancer patients and alternative medicine. N Z Med J 1987; 100: 110-113. Cronan TA, Kaplan RM, Possner I, et al. Prevalence of the use of unconventional remedies for arthritis in a metropolitan community. Arthritis Rheum 1989; 32: 1604-1607. Franklin B. Poor Richard, 1733. In: Labaree LW, editor. The papers of Benjamin Franklin. Vol. 1. January 1706 through December 1734. New Haven: Yale University Press, 1959: 316. Raithatha N. Postmodern philosophy offers a more appropriate system for medicine [letter]. BMJ 1997; 314: 1044. Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine. How to practice and teach EBM. Edinburgh: Churchill Livingstone, 1997. Ivinson AJ. A duty to publish. Nature Med 1998; 4: 1089. Sackett DL, Rosenberg WMC, Gray JAM, et al. Evidence based medicine: what it is and what it isn't. BMJ 1996; 312: 71-72. Herzen A. Epilogue 1849. In: From the other shore. Budberg M, translator. Oxford: Oxford University Press, 1979: 145. Campion EW. Why unconventional medicine? N Engl J Med 1993; 328: 282-283. Eddington A. New pathways in science. London: Cambridge University Press, 1935: 163. Authors' details Royal Perth Hospital, Wellington Street, Perth, WA. Jonathan J Chan, MB BS, Dermatology Registrar. University of Western Australia, Nedlands, WA. Julienne E Chan, BA(Hons), PGDipEd, Postgraduate Scholar. Reprints will not be available from the authors. Correspondence: Dr J J Chan, Department of Medicine, University of Western Australia, 4th Floor G Block, QEII Medical Centre, Verdun Street, Nedlands, WA 6009. jjchanATcyllene.uwa.edu.au Make a comment

Jonathan J Chan · Julienne E Chan

For Debate The hospitalist: a third alternative The role of hospitalist is already evolving in Australia, being filled by Career Medical Officers John M Egan, Mary G T Webber, Michael R D King, Michael Boyd, Gabrielle du Preez-Wilkinson and David Brock MJA 2000; 172: 335-338 The need for the "hospitalist" - The role of the hospitalist - Who best fills this role? - References - Authors' details - - More articles on Administration and health services - More articles on General medicine - More articles on Emergency medicine The hospitalist debate began in the Journal in April 1999, when we published an article by Hillman on how acute-care hospitals are changing. In September 1999, Scott and Phillips suggested that general physicians should take on the role of hospitalist. The start of the debate1 . . . in some countries [there has been] the emergence of a "hospitalist" who has a wide range of expertise, but concentrating more on acute hospital medicine -- more like a general physician, but specialising in acute and serious illness rather than chronic and mainly ambulant medicine. The hospitalist also has advanced resuscitation and procedural skills. They are familiar with the medical comorbidities increasingly associated with surgical patients and understand how different organs fail and interact in acute illness. They are a move back to the generalist physician. The equivalent in Australia is probably the intensive care or emergency physician. The hospitalist also understands about continuity and coordination of patient care, managing the patient's inpatient course and arranging a seamless transition to a community setting. . . . A hospitalist could enable community-based specialists to devote more time to what they do best, rather than being continuously confronted by the dilemma of maintaining a busy professional practice with tight appointment schedules and having seriously ill in-hospital patients who might require their attention day or night in an unpredictable way. Having skilled clinical cover 24 hours a day would also help guarantee patient safety. . . . Australia could explore other ways of achieving the same standards. (Hillman K. MJA 1999; 170: 325-328) The general physician as hospitalist2 . . . patients are more likely to be assured of continuous, integrated and efficient care for a multiplicity of concurrent problems if attended to by general physicians from the time of admission via emergency departments right through to the time of discharge and beyond into ambulatory care. Adequate resourcing of general medical units, greater involvement of general physicians in emergency and intensive care settings, ready access to specialised medical, nursing and allied health expertise as needed . . . (Scott IA, Phillips PA. MJA 1999; 171: 312-314) THERE HAS BEEN AN ONGOING DEBATE in the Journal1-6 on the changing role of the acute-care hospital and, associated with this change, the desirability of a new type of doctor, the hospitalist. It is argued that in the hospital of the near future there will be fewer patients, who will, generally, be more seriously ill than at present.7,8 Leaving aside discussion of the likelihood of this scenario,9 we believe there needs to be a change in the role and experience of doctors who work in this hospital setting: the hospitalist is one suggestion for this change. We argue here that this role is already evolving and currently functioning in a variety of clinical situations in the Australian healthcare system.10The doctor delivering those aspects of seniority, experience and permanence relevant to a hospital generalist is a Career Medical Officer (CMO) (see Box). There are two interrelated parts to this debate: What problems are there with current medical staffing of acute care hospitals? What solutions are available? The need for the "hospitalist" The present model of a Visiting Medical Officer (VMO) who has overall responsibility for the patient and who delegates this responsibility to more junior medical staff in a hierarchical manner (registrars, Resident Medical Officers [RMOs]) while out of the hospital, has a long tradition in medicine. This model, while having some excellent features, may have outlived its usefulness at the beginning of the 21st century. There is increasing evidence of major inadequacies in the functioning of our hospitals.11 This may reflect the way we educate and organise our medical staff. An analysis of the causes of adverse events reported in the Quality in Australian Health Care Study12 showed that human error was involved in 82% of adverse events and that the most common causes were failure in technical performance and cognitive failure: failure to act on available information, failure to consult or investigate, and failure to attend or provide adequate attention. A recent study of critical events (ie, cardiac arrest or unplanned admission to ICU) in an Australian metropolitan teaching hospital13 found that these episodes are frequently preceded by documented clinical instability of the patient and multiple medical review before admission to ICU or cardiac arrest ensued. The authors commented that the patients in their study (generally those with complex medical and surgical conditions) were usually managed initially by the most junior member on the ward (intern or resident). We do not suggest that junior medical officers should not be involved in direct care of the very sick, but there appear to be problems with current hospital practice: the major deficiencies appear to us to be the understandable lack of experience of the junior medical staff and their rapid turnover. This outdated way of organising medical staffing has long caused problems with our nursing colleagues14-16 and others, who assist the overworked and inexperienced medical officers organise their time and energies to most effectively care for and investigate sick patients -- and then repeat this education once again with the next rotation. Is it right to give increasing responsibility to our junior doctors, but leave them at times relatively unsupervised, unsupported and responsible for major medical decisions when their level of training may be inadequate for the task? The Postgraduate Medical Councils in the various states provide increasingly good-quality support for first- and second-year graduates, but this is in a logistical and educational role rather than a directly supportive clinical role. The role of the hospitalist We argue that hospital wards may run more smoothly and there may be greater satisfaction by patients, nursing staff, consultant medical staff and, importantly, junior medical staff if there is rapid access at all times to an onsite experienced medical officer. This senior doctor would be conversant with the dynamics of the ward, have good relationships with and an appreciation of the role of ancillary staff, and have a relatively long-term commitment to the hospital. These attributes would enable him or her to deal with evolving clinical situations before they became major problems. An additional benefit would be the provision of extra educational opportunities for interns and RMOs during their "apprenticeship" years. Hillman1 and Scott and Phillips2 appear to be attempting to address the problems by placing another specialist physician into the increasingly fragmented world of hospital medicine. There are examples of this in the United States,17-19 where the hospitalist is usually (but not always20) a specialist in internal medicine who works predominantly in the hospital setting. Physicians who work outside of the hospital relinquish their responsibility for the patient at the hospital entrance and take it up again on discharge. The unstated but underlying expectation is that the hospitalist would usurp the primary role and responsibility of the attending doctor. We believe this vision of a hospitalist to be fundamentally flawed. On the one hand, it sidelines doctors who should be intimately involved in the inpatient care of patients (including specialist physicians, general practitioners and paediatricians) whose main area of practice remains office based; on the other, it may miss out on providing a "new deal" of care for many patients who are in hospital and whose particular problems might not necessarily fall within the expertise of the intensivist/physician (eg, falls in hospital, paediatric problems, or dementia). Our contention is that this role demands not the narrow focus of the specialist but the broad-based knowledge of the generalist -- someone who can be a "jack of all trades". The preceding contributions to this debate1,2 appear to have as their central vision a hospital filled with medical patients or seriously ill surgical patients who would be better managed by a physician. Most hospitals have, and will continue to have, a much wider range of patients and conditions, including all the major and minor acute problems that one finds in paediatric, gynaecological, obstetric and psychiatric wards. We believe the role calls for a "middle management" doctor who has a breadth of knowledge and experience gained from working in hospitals and who is proficient in as many branches of hospital medicine as possible. The optimal solution is to have someone who is conversant with and experienced in treating seriously ill patients expeditiously, who is quite at home in managing the multiple minor problems that beset hospital patients, and who is used to consulting with a wide range of medical and surgical specialties as needed. The role of hospitalist should be complementary to, not in confrontation with, the established Australian model of inpatient care. That is, it seems to us better to have an experienced doctor "on site" to organise -- not take over -- the management of the hospital inpatient. There is some evidence that hospitalists who have complete control of inpatient care increase, rather than decrease, the length of stay in hospital.21 In the model we propose, the consultant physician, surgeon, paediatrician, gynaecologist, or, increasingly, general practitioner would retain primary responsibility for the patient's management, but would be actively supported by someone who had worked in the hospital system for many years and could competently manage most problems that may arise, at least in the short term. This doctor should also have the trust of, and rapid access to, the consultant staff, and good working relationships with the nursing and paramedical staff (mutual respect of each other's role and abilities) as well as a good understanding of the "mechanics" (eg, layout, routine, and regular practices) of the hospital. Who best fills this role? We believe that Scott and Phillips2 are right in having major reservations about intensivists being responsible for general ward patients. The prevention of a slow deterioration of a general medical or surgical patient to serious illness does not require the considerable skills of an intensivist -- most doctors with experience and education, alerted by protocols that highlight dangerous trends, can quite adequately look after these cases and refer to the intensive care or cardiac care unit if appropriate. The opposite and far more common scenario, that of a relatively minor problem, may well lead to the over-investigation and treatment of a condition that could have been easily handled by a broadly experienced medical officer. Although the general physician may have a better claim to this role (especially rural physicians, who are, by necessity, well-rounded generalists), there is still the problem of a doctor who may be overeducated for some aspects of the work, and undereducated for others. Many of the problems outlined above have in the past been handled by registrars, RMOs and interns. Senior registrars are usually quite able to manage patients without the direct supervision of the VMO; however, this is not necessarily the case with more junior registrars and RMOs. Again, frequent rotation takes well-performing doctors out of the loop just as they attain a level of familiarity and experience with a particular group of patients. Although doctors who performed similar roles had been in the health systems throughout Australia for many years, CMOs were initially brought into service in New South Wales in the early 1980s to maintain experienced medical practitioners in the public hospital system. These doctors were working in posts as unaccredited medical registrars or emergency medical officers, particularly in suburban and rural hospitals. The New South Wales Department of Health10 noted in 1989 that there had been a positive response to the introduction of CMOs: it had increased retention rates of hospital doctors, improved middle grade medical staffing in peripheral hospitals, and addressed the service needs of these hospitals, and the individuals were able to undertake more clinical responsibilities and required less supervision. Many CMOs have now been working in these positions for well over 12 years, and have developed considerable expertise in their area of practice. A recent study in Queensland,22 addressing the training needs and career paths of this cohort of doctors, noted their wide range of practice -- predominantly in emergency medicine ("they make up the majority of the senior emergency work force in Queensland"22), but also in orthopaedics, sexual health, community health, and other areas. The report also pointed out the experience of those who worked in emergency medicine: 73% had worked for more than three years full-time since their third postgraduate year, and 25% had worked for 10 or more years in this capacity. Furthermore, a large proportion (69%) of CMOs had postgraduate qualifications. The study also commented on a major flaw in the Australian Medical Workforce Advisory Committee report The Emergency Medicine Workforce in Australia,23 which totally ignored the role played by CMOs in the staffing of emergency departments. This omission seems to be symptomatic of a "blind spot" by some in the profession to the valuable service provided by these doctors. Many modern private hospitals are turning to CMOs to fill a demand in the medical care of hospitalised patients. In the Sydney area the Hills, Kareena and the Sydney Adventist private hospitals have significant CMO staffing (Dr Stephen Delprado, Deputy Director, Emergency Department, Hills Private Hospital, personal communication), and seven private hospitals in Queensland have CMO cover.22 Most CMOs report that they are quite happy to continue to work in these roles (J M E and M R D K, unpublished survey of 32 rural CMOs in NSW, presented to Directors of Clinical Training meeting, Postgraduate Medical Council, Sydney, May 1996). Recently, some have formed themselves into organised subgroups of the medical workforce. The largest of these (the Career Medical Officers Association) is now taking responsibility for initiating educational and industrial policies for CMOs. Examples of these are the provision of continuing medical education (in association with the Royal College of Pathologists of Australasia); discussions with universities and others about more formal training, qualifications and accreditation; and representations on various committees, including the Hospital Medical Officer Subcommittee of the Medical Training Review Panel. We believe that this broad-based experience and commitment is what makes the CMO the ideal person to take on the role of the hospitalist. Furthermore, the changes necessary to do this are relatively minor and merely a continuation of recent trends in medical workforce utilisation in the modern Australian hospital. In our opinion, CMOs are well able to fulfil such future requirements and do it in the most efficient and cost-effective way. We note the invitation by Scott and Phillips to put the respective views of hospitalist practice to the test in a randomised trial comparing the intensivist or internal medicine models of hospitalist.2 Although we have some misgivings about the applicability of this methodology, we firmly believe that any rigorous evaluation of the various hospitalist models on offer would be incomplete without the inclusion of CMOs. References Hillman K. The changing role of acute-care hospitals. Med J Aust 1999; 170: 325-328. Scott IA, Phillips PA. Hospitals and hospitalists: an alternative view. Med J Aust 1999; 171: 312-314. Denaro CP, Bennett CJ. The changing role of acute-care hospitals [letter]. Med J Aust 1999; 171: 224. Hillman K. Hospitals and hospitalists: an alternative view [letter]. Med J Aust 2000; 172: 299. Sartain JB. Hospitals and hospitalists: an alternative view [letter]. Med J Aust 2000; 172: 299. Phillips PA, Scott IA. Hospitals and hospitalists: an alternative view [letter]. Med J Aust 2000; 172: 299. Komesaroff PA, Clunie GJA, Duckett SJ. What is the future of the hospital system? Med J Aust 1997; 166: 17-22. Braithwaite J. The 21st-century hospital [editorial]. Med J Aust 1997; 166: 6. Braithwaite J, Hindle D. Research and the acute-care hospital of the future. Med J Aust 1999; 170: 292-293. Career Medical Officers (CMOs). Circular no. 89/156. Sydney: NSW Health, 1989. Wilson RMcL, Runciman WB, Gibberd RW, et al. The Quality in Australian Health Care Study. Med J Aust 1995; 163: 458-471. Wilson RMcL, Harrison BT, Gibberd RW, Hamilton JD. An analysis of the causes of adverse events from the Quality in Australian Health Care Study. Med J Aust 1999; 170: 411-415. Buist MD, Jarmolowski E, Burton PR, et al. Recognising clinical instability in hospital patients before cardiac arrest or unplanned admission to intensive care. A pilot study in a tertiary-care hospital. Med J Aust 1999; 171: 22-25. Lublin J, Gething L. RNs as teachers of junior doctors. Aust J Advanced Nursing 1992; 10(2): 3-9. Agnew T. Just rewards on the wards. Nursing Times. 1995; 91(34): 19. Junior doctors turn to nurses for help. Nursing Standard 1999; 13(47): 8. Wachter RM. An introduction to the hospitalist model. Ann Intern Med 1999; 130: 338-342. Sox HC. The hospitalist model: perspectives of the patient, the internist, and internal medicine. Ann Intern Med 1999; 130: 368-372. Schroeder SA, Schapiro R. The hospitalist: new boon for internal medicine or retreat from primary care? Ann Intern Med 1999; 130: 382-387. NAIP affiliates with the American College of Physicians. <http://www.naipon line.org/hist.htm>. Accessed 2 March 2000. Jackson JL. The international experience with hospitalists. The Hospitalist 1997; Summer. Available at <http://www.naiponline.org/archives/guest.htm>. Accessed 2 March 2000. Bricknall B, Daly M, Catchpole M. The career paths, training needs and future role of non-specialist senior medical officers in the Queensland public health care system. An exploratory study. Brisbane: Health Advisory Unit, Queensland Health, 1999. Australian Medical Workforce Advisory Committee. The Emergency Medicine Workforce in Australia. Sydney: AMWAC, 1997. Summary available at <http://amwac.health.nsw.gov.au/corporate-services/amwac/emerg.htm>. Accessed 2 March 2000. Authors' details Goulburn Base Hospital, Goulburn, NSW. John M Egan, MB BS, Career Medical Officer, Emergency Department. Kareen Private Hospital, Sydney, NSW. Mary G T Webber, MB BS, Career Medical Officer, Emergency Department, and President Career Medical Officers Association. Coffs Harbour Base Hospital, Coffs Harbour, NSW. Michael R D King, FRACGP, FACRRM, Director of Emergency Services. Camden Hospital, Sydney, NSW. Michael Boyd, MB BS, Co-ordinator of Emergency Department. Prince Charles Hospital, Brisbane, QLD. Gabrielle du Preez-Wilkinson, FRACMA, AFCHSE, Medical Officer, Emergency Department. Tweed Heads Hospital, Tweed Heads, NSW. David Brock, MB BS, Career Medical Officer, Emergency Department. Reprints: Dr J M Egan, PO Box 131, Goulburn, NSW 2580. eganjATinteract.net.au Make a comment What is a Career Medical Officer? Officially, a Career Medical Officer (CMO) is a grade of medical officer employed by the New South Wales Department of Health. Unofficially, and more accurately, CMOs are: "middle management" doctors who increasingly perform in responsible and demanding clinical roles; doctors beyond the second postgraduate year and working in clinical medicine; not general practitioners (although some work in both roles), nor specialists, nor in training for these roles; a distinct subgroup within the wider medical community, with their own aspirations, experience and educational needs. In different parts of Australia doctors in this role have different designations: Senior Medical Officer (SMO: Qld, WA, SA, NT); Hospital Medical Officer (HMO: Vic); Career Medical Officer (CMO: NSW, but this title also known and occasionally used in other states). In practice, awards for these medical officers range from registrar to staff specialist range. How many CMOs are there? No accurate numbers are available. Recent medical workforce surveys show 652 "other hospital career" doctors in NSW. In Queensland, there are 149 funded SMO positions. These figures are almost certainly an underestimate as they do not take into account community CMOs who may be labelled as GPs or specialists, and self-reporting of some hospital CMOs as registrars or staff-specialists. Likely "ball-park" figures are 1000 in NSW, and 2000 Australia-wide, but there may be significantly more. Where do CMOs work? About 60%-70% of CMOs work in emergency departments in rural, suburban and private hospitals. CMOs also work in psychiatry, sexual health, women's health, police forensic, intensive care, neonatal, orthopaedics, and other areas. In rural Queensland and NSW, CMOs are the predominant senior doctors in emergency departments. What education do CMOs have? No formal qualifications are required at present, but at least half have postgraduate qualifications. Many (especially in emergency departments) have early management of severe trauma (EMST), emergency life support (ELS), or advanced paediatric life support (APLS) qualifications. Discussions are in progress with the University of Newcastle regarding more formal postgraduate qualifications for CMOs. The Postgraduate Medical Councils may have an advisory or other role in CMO education and training. The Career Medical Officers Association (CMOA) website <http://www.cmoa.ican.net.au/> has details of current interest for CMOs and others. The information in this Box comes from many sources, including the NSW Medical Labour Force Annual Survey 1998, the CMOA website, the CMOA database, Bricknall et al22 and discussions with CMOs. Back to text

John M Egan · Michael Boyd · David Brock

Clinical Update Genetic testing for Alzheimer's disease Peter K Panegyres, Jack Goldblatt, Ian Walpole, Carmela Connor, Toni Liebeck and Karen Harrop MJA 2000; 172: 339-343 Abstract - Recommendations for gene testing in Alzheimer's disease - Conclusions - Acknowledgements - References - Authors' details - - More articles on Genetics Abstract Genetic factors are important in the development of Alzheimer's disease (AD). Familial AD can result from rare mutations in some genes. Other genes, such as the apolipoprotein E gene (APOE), operate as risk factors for late-onset sporadic AD. On a background of advances in the genetics of AD we suggest a way in which genetic information may be used in the diagnosis of AD. If there is a positive family history of early-onset dementia and the clinical features suggest AD, patients may be tested for presenilin and amyloid precursor protein gene mutations with appropriate pretest and post-test counselling. Predictive testing should be performed under guidelines developed by the World Federation of Neurology and the Human Genetics Society of Australasia. The usefulness of APOE genotyping as an adjunct to conventional diagnostic tests is unknown; data suggest it has low sensitivity and specificity and may have little predictive value in an individual patient. APOE genotyping should not be performed in asymptomatic individuals, except as part of an ethically approved research project; this recommendation is supported by a number of international consensus statements. APOE testing should not be used as a diagnostic test without adequate pretest and post-test counselling, education and support. APOE testing should not be used as a sole diagnostic test in the work-up of patients with AD. Genetic risk factors other than APOE require validation and should not be used routinely, except as part of an ethically approved research protocol. Alzheimer's disease (AD) is one of the major healthcare problems facing First World countries. In 1995, 130 000 Australians aged over 65 years had moderate to severe dementia, and by 2041 the number of people with dementia in Australia is expected to increase by 254%.1 In recent years there have been major advances in the elucidation of genetic factors in both familial and sporadic AD. Unfortunately, the accumulation of this genetic information has outpaced understanding among the medical and genetic communities of the most appropriate way it can be used clinically. This has led to diagnostic kits for DNA markers having to be withdrawn because of misuse in counselling individuals about future risks.2 Consumer-led demand for diagnostic tests and pressure from companies that make them raise major ethical considerations about the role of genetic testing in managing families at possible risk. These developments have encouraged us to develop evidence-based recommendations on the use of molecular genetic testing in Alzheimer's disease. Background information on the genetics of Alzheimer's disease is provided in Box 1. Recommendations for gene testing in Alzheimer's disease Diagnosis of dementia The diagnosis of AD requires assessment by a clinician skilled in diagnosing dementia, using the criteria established by the National Institute of Neurological and Communicative Disorders and Stroke - Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA),4 shown in Box 2. The diagnosis can sometimes be difficult, and expertise is required to distinguish AD clinically from other disorders such as frontotemporal atrophy and prion diseases, especially in younger patients. The diagnostic work-up of the patients is best performed in a facility with counselling and support staff to help patient and carer cope with the diagnosis of dementia. Collaboration with a neuropathology laboratory can enhance such a service by providing postmortem confirmation of the diagnosis. Education, counselling, support A multidisciplinary team, including a neurologist or specialist physician, neuropsychologists, social workers, allied health workers (occupational therapists, speech therapists), all working together with the patient's general practitioner, helps patients and their carers understand the diagnosis of AD and its implications. Contact with a caring, multidisciplinary team can support the patient, carer, and family in crises such as the development of intercurrent medical problems requiring hospitalisation and respite for patient or carer. Such a team can advise the patient, carer and family on the suitability and appropriateness of genetic testing. Genetic testing If genetic testing is considered then the staff of the multidisciplinary team must know the implications, risks and limitations of the proposed tests and counsel patients, carers and families accordingly. They must have the expertise to counsel patients, carers and families about psychosocial implications, confidentiality, and issues related to employment and insurability, of the genetic tests requested. Accredited laboratory and DNA result disclosure The laboratory that tests the DNA specimen must be accredited by the National Association of Testing Authorities, Australia, which advises on specimen handling, the maintenance of strict confidentiality, and good laboratory practice (Box 3). The DNA result should be given to the clinician who requested the test, who will disclose the result to the patients and carer in strictest confidence with the help of counsellors. As has been shown with Huntington's disease,34 follow-up by counsellors will help to decrease adverse reactions such as suicide, attempted suicide and psychiatric hospitalisation. Familial early-onset AD If there is a positive family history of early-onset dementia, and pedigree analysis suggests autosomal dominant AD, the patient and family should be referred to a clinician with an interest in familial dementia for confirmation of the diagnosis. The patient and family should then be managed by a multidisciplinary team of experts in genetic neurodegenerative disorders (such as Huntington's disease) and predictive gene testing. Genetic testing should only be offered in a comprehensive, structured, clinico-laboratory program where mutations in PS1, PS2 and APP would be sought in affected individuals. Gene testing is not recommended for sporadic cases of early-onset AD without a definite family history. Predictive testing in unaffected and asymptomatic individuals from families in which causative mutations have been discovered must follow guidelines as developed for Huntington's disease.5-8 Only about 6% of patients at risk of Huntington's disease request the gene test, probably because many at-risk people decide against the test once they receive full information of its implications.35 The likelihood of suicide, attempted suicide or psychiatric hospitalisation after predictive testing is no greater than in the general population with symptomatic Huntington's disease, and this is probably the result of good counselling and support.34 Similar considerations may apply to AD. Like Huntington's disease, AD is an incurable condition with devastating consequences, and there are ethical issues (such as patients not wanting to know, and implications for employment and insurability) relating to predictive gene testing in such situations. These ethical issues probably contribute to the low uptake of testing for Huntington's disease and will probably be relevant to AD also. Other dilemmas in predictive testing for AD relate to performing tests in individuals with 25% risk when an unaffected or undiagnosed parent does not request a gene test; a positive result in such individuals would result in an unwanted gene result for the parent. This represents a difficult situation for predictive gene testing programs. Similar problems arise in twins if only one wants to be tested. For ethical reasons, as in Hungtington's disease, children should not be tested for AD. DNA banking should be considered for individuals with a family history of AD who may not want a test at present, or who may not have any of the recognised mutations -- future testing could be carried out if other mutations are recognised. Sporadic AD In patients with the clinical diagnosis of sporadic AD, gene testing for APOE ε4 status or other genetic factors is not recommended. The clinical usefulness of these tests has not been established, and there is no evidence that they improve the sensitivity and specificity of the clinical diagnosis of AD sufficiently to alter the standard diagnostic work-up of these patients. The APOE ε4 genotype should never be used as a sole diagnostic test for the diagnosis of AD. Conclusions Our recommendations are summarised in Box 4. Acknowledgements A National Health and Medical Research Council fellowship awarded to Dr Panegyres supported this work. References Henderson AS, Jorm AF. Dementia in Australia. Aged and Community Care Service Development and Evaluation Report No. 35. Canberra: AGPS, 1998. Lehrman S. Genetic testing for Alzheimer's disease "not appropriate". Nature 1997; 389: 898. Tanzi RE, Kovacs DM, Kim T-W, et al. The gene defects responsible for familial Alzheimer's disease. Neurobiol Dis 1996; 3: 159-168. McKhann G, Drachman DD, Folstein M, et al. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of the Department of Health and Human Services Task Force in Alzheimer's disease. Neurology 1984; 34: 939-944. World Federation of Neurology: Research Committee. Research Group on Huntington's Chorea. Ethical issues policy statement on Huntington's disease molecular genetics predictive test. J Neurol Sci 1989; 94: 327-332. Went L. Ethical issues policy statement on Huntington's disease molecular genetics predictive test. International Huntington Association. World Federation of Neurology. J Med Genet 1990; 27: 34-38. Guidelines for the molecular genetics predictive test in Huntington's disease. International Huntington Association (IHA) and the World Federation of Neurology (WFN) Research Group on Huntington's Chorea. Neurology 1994; 44: 1533-1536. Walpole I, Bankier A, Blackwell J, et al. Guidelines for DNA predictive testing. Bull Hum Genet Soc Austral 1998; 11: 10-14. Goate AM, Chartier-Harlin MC, Mullan MC, et al. Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer's disease. Nature 1991; 349: 704-706. Blacker D, Tanzi RE. The genetics of Alzheimer disease. Arch Neurol 1998; 55: 294-296. Levy-Lehad E, Wasco W, Podrkaj P, et al. Candidate gene for the chromosome 1 familial Alzheimer disease locus. Science 1995; 269: 973-977. Saunders AM, Strittmatter WJ, Schmechel D, et al. Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease. Neurology 1993; 43: 1467-1472. Strittmatter WJ, Saunders AM, Schmechel D, et al. Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease. Proc Natl Acad Sci USA 1993; 90: 1977-1993. Gomez-Isla T, West HL, Rebeck GW, et al. Clinical and pathological correlates of apolipoprotein E ε4 in Alzheimer's disease. Ann Neurol 1996; 39: 62-70. Jonker C, Schmand B, Lindeboom J, et al. Association between apolipoprotein E ε4 and the rate of cognitive decline in community-dwelling elderly individuals with and without dementia. Arch Neurol 1998; 55: 1065-1069. Hyman BT, Gomez-Isla T, Briggs M, et al. Apolipoprotein E and cognitive change in an elderly population. Ann Neurol 1996; 40: 55-66. Mayeux R, Saunders AM, Shea S, et al. Utility of the apolipoprotein E genotype in the diagnosis of Alzheimer's disease. N Engl J Med 1998; 338: 506-511. McConnell LM, Sanders GD, Owens DK. Evaluation of genetic tests: APOE genotyping for the diagnosis of Alzheimer's disease. Genet Test 1999; 3: 47-53. American College of Medical Genetics and American Society of Human Genetics Working Group on APOE and Alzheimer Disease. Statement on use of apolipoprotein E testing for Alzheimer disease. JAMA 1995; 274: 1627-1629. McConnell LM, Koenig BA, Greely HT, Raffin TA, and Alzheimer Disease Working Group of the Stanford Programme in Genomics, Ethics and Society. Genetic testing and Alzheimer disease: Has the time come? Nature Medicine 1998; 4: 757-759. National Institute on Aging and Alzheimer's Association Working Group. Apolipoprotein E genotyping in Alzheimer's disease. Lancet 1996; 347: 1091-1095. Post SG, Whitehouse PJ, Binstock RH, et al. The clinical introduction of genetic testing for Alzheimer disease. JAMA 1997; 277: 832-836. Relkin NR, Kwon YJ, Tsai J, Gandy S. The National Institute on Aging/ Alzheimer's Association recommendations on the application of apolipoprotein E genotyping to Alzheimer's disease. Ann NY Acad Sci 1996; 802: 149-171. Jobst KA, Hindley NJ, King E, Smith AD. The diagnosis of Alzheimer's disease: a question of image? J Clin Psychiatry 1994; 55 Suppl: S22-S31. Spinnler H, Della Sala S. The role of clinical neuropsychology in the neurological diagnosis of Alzheimer's disease. J Neurol 1988; 235: 258-271. Zakzanis KK. Quantitative evidence for neuroanatomic and neuropsychological markers in dementia of the Alzheimer's type. J Clin Exp Neuropsychol 1998; 20: 259-269. Bullido MJ, Artiga MJ, Recuero M, et al. A polymorphism in the regulatory region of APOE associated with risk for Alzheimer's dementia. Nat Genet 1998; 18: 69-71. Blacker D, Wilcox MA, Laird NM, et al. Alpha-2 macroglobulin is genetically associated with Alzheimer disease. Nat Genet 1998; 19: 357-360. Kang DE, Saitoh T, Chen X, et al. Genetic association of the low density lipoprotein receptor-related protein gene (LRP), an apolipoprotein E receptor, with late-onset Alzheimer's disease. Neurology 1997; 49: 56-61. Montoya SE, Aston CE, Dekosky ST, et al. Bleomycin hydrolase is associated with risk of sporadic Alzheimer's disease. Nat Genet 1998; 18: 211-212. Lehmann DJ, Johnston C and Smith AD. Synergy between the genes for butyrylcholinesterase K variant and apolipoprotein E4 in late onset confirmed Alzheimer's disease. Hum Mol Genet 1997; 11: 1933-1936. Panegyres PK, Mamotte CDS, Vasikaran SD, et al. Butyrylcholinesterase K variant and Alzheimer's disease. J Neurol 1999; 246: 369-370. Payami H, Schellenberg GD, Zareparsi S, et al. Evidence for association of HLA-A2 allele with onset age of Alzheimer's disease. Neurology 1997; 49: 512-518. Almqvist EW, Bloch M, Brinkman R, et al. On behalf of an international Huntington disease collaborative group. A worldwide assessment of the frequency of suicide, suicide attempts, or psychiatric hospitalization after predictive testing for Huntington's disease. Am J Hum Genet 1999; 64: 1293-1304. Taylor SD. Demand for predictive genetic testing for Huntington's disease in Australia, 1987 to 1993. Med J Aust 1994; 161: 351-354. (Received 2 Aug 1999, accepted 14 Feb 2000) Authors' details Neurosciences Unit, Health Department of Western Australia, Perth, WA. Peter K Panegyres, PhD, FRACP, Neurologist, and NHMRC Fellow, Department of Neuropathology, Royal Perth Hospital. Carmela Connor, MPsychol, Senior Clinical Psychologist. Toni Liebeck, BSW, Senior Social Worker. Genetic Services of WA, King Edward Memorial Hospital for Women, Perth, WA. Jack Goldblatt, MD, FRACP, Director. Ian Walpole, MB BS, FRACP, Consultant Geneticist. Karen Harrop, BSc, Genetic Counsellor. Reprints: Dr P K Panegyres, Department of Neuropathology, Royal Perth Hospital, Wellington Street, Perth, WA 6000. peter.panegyresATrph.health.wa.gov.au Make a comment 1: The genetics of Alzheimer's disease Familial early-onset Alzheimer's disease Studies of families in which Alzheimer's disease (AD) was inherited in an autosomal dominant pattern led to the discovery of three pathogenic loci that account for about 50% of all cases of early-onset AD3 (Table). As the condition is heterogeneous, every family with early-onset AD should be offered investigation as part of a coordinated DNA testing program for neurological disease. Mutations in the amyloid precursor protein gene (APP) were the first mutations related to early-onset AD,9 and account for 10%-20% of familial AD.10 Two presenilin genes, PS1 and PS2, are also associated with early-onset familial AD -- almost 50% of cases result from mutations in PS1,10 while mutations in PS2 are rare.11 As reproducibility in PS1 mutation testing has not been established in some laboratories, caution is warranted. Implications for genetic testing: Gene testing for early-onset AD is probably best performed in the context of well-designed, ethically approved research projects involving large families with clear documentation in multiple-affected members who inherited the condition in an autosomal dominant fashion. In some Australian centres, patients with early-onset AD are routinely tested outside of research protocols. As mutations in the presenilin and amyloid precursor protein genes do not account for all cases of early-onset AD, negative screening results for these mutations in an affected individual would not exclude a genetic cause of the disease. Sporadic late-onset Alzheimer's disease APOE: Over 90% of patients with AD have no family history of the condition. One of the more important discoveries in the understanding of these sporadic late-onset cases was that a polymorphism of the apolipoprotein E gene (APOE) was a risk factor (Table).12,13 APOE has three alleles, designated ε2, ε3 and ε4. The ε4 allele is associated with AD in 20%-30% of the general population and in 45%-60% of patients with AD.14 The homozygous genotype, APOE ε4/ε4, is found in 12%-15% of patients with AD, but in only 2%-3% of the general population.14 While not everyone homozygous for APOE ε4 develops dementia, having this genotype might increase the chance of AD developing at an earlier age.15 Approximately 30% of people homozygous for APOE ε4 develop AD.16 The odds ratio for this, based on analysis of 1899 patients aged over 65 years, is 1.37 (versus 0.53 for the APOE ε2 allele).16 In this same study, the age-adjusted odds ratio for incident dementia in individuals homozygous for the ε4 allele was 1.89, and 25% of cognitively normal subjects had at least one ε4 allele. Further, absence of an ε4 allele does not prevent the development of dementia and AD, and 85% of elderly people with the APOE ε4/ε4 genotype did not have evidence of cognitive decline.16 In a pathologically proven series, a single APOE ε4 allele had a sensitivity of 65% and a specificity of 68% for the diagnosis of AD.17 When used with conventional clinical criteria, APOE ε4 testing might increase the diagnostic sensitivity and specificity by 5%-10%; therefore, its role requires further validation.15,17 Implications for genetic testing: APOE ε4 genotyping should not be used in the routine assessment of patients with suspected dementia, as it does not add significant information to other diagnostic investigations such as computed tomography (CT) and neuropsychological assessment.18 If the DNA test is performed it should not be done without adequate pretest counselling as to its limitations and implications, or without adequate post-test psychosocial support. The results need to be stored confidentially in view of the implications for other, unrelated conditions (eg, APOE allele status was used to predict risk in cardiovascular disease long before its significance in AD was known), insurability, employment and psychosocial coping for affected individuals and at-risk families. Thus, APOE genotyping should only be performed as part of a well-structured, ethically approved research study investigating issues about the role of APOE in the pathogenesis of AD. The evidence does not support using APOE e4 genotyping as a predictive test for the development of AD, as the exact significance of an APOE ε4 allele in asymptomatic individuals has not been confirmed.19-23 APOE ε4 genotyping should not be used as a sole diagnostic test for AD. Diagnosis requires specialist referral for investigations, such as a CT scan (which has a 94% positive predictive value24) and neuropsychological tests (85%-90% positive predictive value for the diagnosis of dementia, with less than 5% overlap of neuropsychology scores between patients and controls25,26). A positive APOE ε4 test is not diagnostic of Alzheimer's disease, as a single APOE ε4 allele has a positive predictive value of 65% and a negative predictive value of 68%. The presence of the APOE ε4 allele does not exclude other causes of dementia. For example, a 1998 study showed that about 5% of patients with clinical criteria for the diagnosis of AD were homozygous for APOE ε4, but did not have pathological features of AD.17APOE ε4 diagnostic kits should not be used in the clinical assessment of dementia. (Although such kits were previously available in the United States, they had to be withdrawn because of misuse.2) Other genetic factors: A number of genetic factors in addition to APOE ε4 have been associated with sporadic AD (Table). These include APOE A/T polymorphism in the promoter region,27α2 macroglobulin 5' splice site deletion on exon 18,28 low-density lipoprotein-receptor-related protein,29 the G/G homozygous state of the bleomycin hydrolase gene,30 butyrylcholinesterase K variant,31,32 and the major histocompatibility A2 antigen.33 The contribution of these factors to the diagnosis of AD requires more research, as they have not been sufficiently validated to be used routinely. Implications for genetic testing: These genetic factors need confirmation and further analysis as to their role in the diagnosis of AD and should not be used as diagnostic or predictive tests outside of research programs. Clinical application of genetic factors in Alzheimer's diseaseChromosomeDiagnostic testing*Predictive testing†Pathogenic loci Presenilin 1 (PS1)14++ Presenilin 2 (PS2)1++ Amyloid precursor protein (APP)21+ + Risk loci Apolipoprotein E (APOE ε4)19± - Apolipoprotein E -491AA19- - α2 Macroglobulin12-- Low-density receptor-related protein12- - Bleomycin hydrolase17-- Butyrylcholinesterase K variant3-- *In symptomatic individuals with clinical evidence of autosomal dominant familial or sporadic Alzheimer's disease, using NINCDS-ADRDA criteria for the diagnosis of Alzheimer's disease.4 †In asymptomatic individuals using guidelines as developed for Huntington's disease.5-8 -491AA=A/A polymorphism at position -491 in the transcription regulation region of APOE. Back to text 2: National Institute of Neurological and Communicative Disorders and Stroke - Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) criteria for the clinical diagnosis of Alzheimer's disease4 I. The criteria for the clinical diagnosis of probable Alzheimer's disease include: Dementia established by clinical examination and documented by the Mini-Mental Test, Blessed Dementia Scale, or some similar examination, and confirmed by neuropsychological tests; Deficits in two or more areas of cognition; Progressive worsening of memory and other cognitive functions; No disturbance of consciousness; Onset between ages 40 and 90, most often after age 65; and Absence of systemic disorders or other brain diseases that could account for the progressive deficits in memory and cognition. II. The diagnosis of probable Alzheimer's disease is supported by: Progressive deterioration of specific cognitive functions such as language (aphasia), motor skills (apraxia), and perception (agnosia); Impaired activities of daily living and altered patterns of behaviour; Family history of similar disorders, particularly if confirmed neuropathologically, and laboratory results showing: normal lumbar puncture as evaluated by standard techniques, normal pattern or non-specific changes in an electroencephalogram, such as increased slow-wave activity, and evidence of cerebral atrophy on computed tomography, with progression documented by serial observation. III. Criteria for diagnosis of definite Alzheimer's disease are: The clinical criteria for probable Alzheimer's disease, and Histopathological evidence from a biopsy or autopsy. Back to text 3: Australasian centres where genetic testing for presenilin, APP mutations and APOE genotyping and counselling is available Applied Molecular Biology Unit Biochemistry State Health Laboratory Brisbane, QLD Department of Pathology Royal Brisbane Hospital, Brisbane, QLD Molecular Pathology Laboratory Sullivan Nicolaides Pathology Taringa, QLD Laboratory and Community Genetics Kolling Institute of Medical Research Royal North Shore Hospital St Leonards, NSW Institute of Medical and Veterinary Science [IMVS] Adelaide, SA The Neurosciences Unit, Health Department of Western Australia, and Department of Neuropathology, Royal Perth Hospital, Perth, WA Hollywood Private Hospital Perth, WA Molecular Pathology Laboratory Canterbury Health Laboratories Christchurch, New Zealand Back to text 4: Recommendations for genetic testing in Alzheimer's disease Mutation testing for abnormalities in PS1, PS2 and APP should only be considered where the family history is appropriate and by referral to appropriate comprehensive, predictive testing programs. The absence of known mutations does not protect against the development of other types of dementia. The use of APOE ε4 genotyping as an adjunct to conventional diagnostic measures is unknown and the data suggest that it has low sensitivity and specificity and is of little diagnostic value in an individual patient. It has low predictive value in asymptomatic individuals and its use in this situation should be discouraged, except in well-defined research protocols with appropriate institutional ethics approval. APOE ε4 genotyping should not be used as the sole diagnostic test in patients in whom AD is suspected on clinical grounds. APOE ε4 genotype testing should not be offered without adequate pre-test and post-test counselling, education and support in patients in whom AD is suspected on clinical grounds. Genetic testing should follow carer or patient consent. APOE ε4 genotype analysis should not be performed in asymptomatic individuals. This position is supported by a number of overseas consensus statements.19-23 Other DNA-based genetic risk factors should not be used in the routine assessment of patients, except as part of an ethically approved research protocol. Back to text

Peter K Panegyres · Jack Goldblatt · Ian Walpole · Carmela Connor · Toni Liebeck · Karen Harrop