Volume 172 Issue 1

Medicine and the community Patient attitudes to commonly promoted medical interventions Stephen P Fitzgerald and George Phillipov MJA 2000; 172: 9-12 For editorial comment, see Bastian Introduction - Methods - Data sources - Questionnaire - Design and settings - Statistical analysis - Results - Discussion - References - Authors' details - - More articles on Social issues Abstract Objective: To survey attitudes about three "best practice" medical interventions (hormone replacement therapy [HRT], thrombolysis for acute myocardial infarction [THROM] and coronary artery by-pass surgery [CABS]) in a sample of patients, and identify factors associated with those attitudes. Settings: Metropolitan tertiary care hospital outpatient clinics (survey 1, April 1997), two general practice surgeries (survey 2, May 1997), and one general practice surgery (survey 3, October 1997). Design: Patients completed a questionnaire while waiting for their clinical consultation. Attitude scores were measured on an 11-category Likert scale ranging from - 5 (definitely would not) to + 5 (definitely would) for acceptance of proposed medication or surgery. Participants: 85 (participation rate, 85%), 77 (94%) and 95 (97%) in surveys 1, 2 and 3, respectively. Surveys 1 and 2 constituted the primary study group (n = 162). Patients aged ≥50 years or reporting heart disease were excluded from the HRT analyses; patients aged ≥65 years were excluded from the THROM and CABS analyses. Results: The median attitude scores for HRT (n = 58), THROM and CABS (n = 111) were - 2.95 (95% CI, - 5 to - 2.1), - 0.5 (95% CI, - 0.9 to 0) and - 0.1 (95% CI, - 0.5 to + 1.3), respectively. Decreasing the risk-benefit ratio fourfold for HRT in survey 3 (n = 68) increased the median score to - 0.75 (95% CI, - 2.3 to 0). Conclusions: Patients do not view favourably the risk-benefit ratio of the three surveyed medical interventions. These attitudes may present a major impediment to most primary prevention programs. Introduction In recent years, evidence-based medicine has been proposed as a fundamental advance in the delivery of "best practice'' medicine.1 Accordingly, an implicit requirement exists for physicians to be aware of the best evidence, which has been suggested to be the outcomes of randomised clinical trials or the meta-analyses of such trials.2However, consensus has also developed that patient empowerment, through complete specification of disease and treatment risks as well as likely outcome benefits, should be standard in any decision-making process.3 While evidence-based medicine may lead to a clear preference by doctors for treatment, acceptance of this same preference by patients, when apprised of all the relevant risks and benefits, has, to our knowledge, received limited study.4,5 Moreover, Schwartz et al6 have presented findings which question the ability of many patients to comprehend elementary probability concepts, understanding that is crucial to any goal of patient empowerment. There are also reports which suggest that the necessity for proper, informed choice has been compromised,7 true risk odds overstated8 or biased epidemiological data presented9 to ensure patient participation. Such reports infer that patients may not necessarily perceive, or be able to perceive, the risk-benefit ratio for many treatments as favourably as their physicians. The purpose of our study was to analyse the questionnaire-based attitudes of patients on the risks and benefits of a number of prominently promoted medical interventions.10-12 Methods Data sources The risk-benefit data for the questionnaires were based on the published results of trials for hormone replacement therapy (HRT),13-15 thrombolysis for acute myocardial infarction (THROM),16 and coronary artery bypass surgery (CABS).17 Questionnaire The results of published trials were reported as an absolute (rather than relative) percentage of the likely benefit or adverse outcome, to reduce any potential bias with respect to patients' interpretation of results.18-21 To minimise response bias due to specific diseases, the questionnaire statements were generically worded; participants did not know the name of the therapy. Participants were asked to indicate their preference on an 11-category Likert scale ranging from "definitely would not" to "definitely would" accept the treatment (medication or surgery). For analysis, responses were coded from - 5 to + 5, with zero representing uncertainty. The survey participants were also given an option to provide free-text comments for their responses. These were subsequently categorised into 10 general themes: the age at which intervention occurs; prior experience of the situation (either personal or acquaintances); the odds of success offered by the intervention; an aversion to taking any form of risk; a belief that life-events are predetermined; full confidence in doctors' advice; prefer alternative medicine procedures; insufficient information to make decision; a willingness to take any risk for better health; and all other reasons Design and settings The first survey was conducted at the Queen Elizabeth Hospital's endocrine and diabetes outpatient clinics (April 1997), and was coordinated by the nursing staff. Queen Elizabeth Hospital is a metropolitan, university-affiliated teaching hospital. The second survey was conducted at two general practice surgeries, which each received 50 questionnaires (May 1997). Each surgery's receptionist handled patient recruitment. The survey used the same questionnaire as the first survey, but without the "no difference" effect phrase, to assess whether this phrase had had an overly negative bias on patient scores. After preliminary assessment of the two surveys, a third was conducted (October 1997) at a single general practice surgery, to assess the effect of a marked decrease in the risk-benefit ratio on patients' attitude towards HRT. In all three surveys, patients were approached and asked to complete a questionnaire while waiting for their clinical consultation. One hundred questionnaires were made available for each survey, and each survey was conducted over a two-week period. There were no exclusion criteria for survey recruitment. The main reason for patient refusal was poor reading skills. The demographic and other details of participants are shown in Box 1. As the goal of our study was to assess patient attitudes in a prospective manner, age exclusion criteria (and cardiovascular disease comorbidity for HRT) consistent with the proposed intervention were applied before analysis. The study was approved by the hospital ethics committee. Statistical analysis Non-parametric statistical tests were performed using SPSS 6.1.22 Confidence intervals for medians and differences were calculated as previously described.23 Results On an 11-point Likert scale, where - 5 indicates "definitely would not" and +5 indicates "definitely would" accept the treatment, the median response for HRT in surveys 1 and 2 was - 2.95 (95% CI, - 5 to - 2.1). In survey 3, with a fourfold decrease in risk-benefit, this was significantly increased to - 0.75 (95% CI, - 2.3 to 0). For THROM, the median response was - 0.5 (95% CI, - 0.9 to 0), and for CABS, - 0.1 (95% CI, - 0.5 to + 1.3). For further details, see Boxes 2 and 3. Discussion Our results indicate that a discrepancy in perceived benefit exists between our participants and that of the medical community towards certain common medical interventions. In none of the responses to the questions posed was there a median score that signified acceptance of the proposed risk-benefit for the intervention. Our findings suggest that patients' perceptions are more influenced by the situation and risk than the odds of successful outcome. Thus, asymptomatic patients offered primary prevention for heart disease (HRT) had a very negative attitude, but viewed THROM and CABS treatment more favourably. This may reflect that patients consider these latter situations as potentially life-threatening and are therefore more willing to seek benefit (life) as a matter of urgency (see Box 3 free-text responses). Alternatively, it might be argued that the change in score relates to the absence of cancer as a potential side-effect, given the alarm this disease engenders.24 We were able to demonstrate that lowering the risk-benefit for HRT led to a less negative attitude. Given the magnitude in risk-benefit reduction (fourfold) we consider the overall change modest, since even at this degree of risk-benefit the attitude towards HRT was still negative. We therefore believe that our findings are essentially robust with respect to moderate changes in risk-benefit. A limitation of our study was the small sample size, which decreased the statistical power to detect differences. The external generalisability of the findings is also limited by a lack of information on participants' socioeconomic status and level of education. However, the similarity of attitude scores across the different settings infers a degree of robustness towards such confounders. The favourable attitude towards CABS by patients with existing heart disease is consistent with a recent report that patients actively seek aggressive post-myocardial infarction treatment from their doctors.25 The difference between the findings for THROM and CABS may reflect greater public awareness and expectations concerning CABS treatment. It may be argued that our results, which show an overall equivocal attitude towards "best-practice" guidelines, are confounded by patients' inability to adequately interpret the probabilistic concepts.6 However, about 75% of patients providing written responses to the HRT questions indicated a clear aversion to any risk-taking; this same concern, although diminished, was also described for THROM and CABS. Moreover, for THROM and CABS, while a significant proportion of patients claimed that their decision was influenced by the odds, the actual median scores suggest the odds were viewed unfavourably. A fundamental problem with evidence-based medicine, therefore, is that it encourages practices by physicians which may often conflict with the evolving paradigm of shared decision-making and even their own value judgements with respect to the risks and benefits of certain treatments. This is most likely to occur for preventive treatments, a medical strategy least favoured by our patients. It is also important to acknowledge that a decision by a patient that is different from that advocated by their physician does not imply the patient does not understand the information.26 In conclusion, our findings show that patients have a variety of reasons for accepting or rejecting particular medical strategies, and, even where those reasons are based directly on risk-benefit outcomes, they are seldom as positive as proponents of evidence-based medicine might expect. Accordingly, widespread promulgation and instigation of "best-practice" guidelines for the medical interventions evaluated in this study would appear limited by any increasing patient role in the decision-making process.27 References Evidence-Based Medicine Working Group. Evidence-based medicine. A new approach to teaching the practice of medicine. JAMA 1992; 268: 2420-2425. National Health and Medical Research Council. A guide to the development, implementation and evaluation of clinical practice guidelines. Canberra: Commonwealth of Australia, 1999. Available at <http://www.nhmrc.health.gov.au/publicat/synopses/cp30syn.htm>. Bogardus Jr ST, Holmboe E, Jekel JF. Perils, pitfalls, and possibilities in talking about medical risk. JAMA 1999; 281: 1037-1041. McNeil BJ, Weichselbaum R, Pauker SG. Fallacy of the five-year survival in lung cancer. N Engl J Med 1978; 299: 1397-1401. McNeil BJ, Weichselbaum R, Pauker SG. Speech and survival: tradeoffs between quality and quantity of life in laryngeal cancer. N Engl J Med 1981; 305: 982-987. Schwartz LM, Woloshin S, Black WC, Welch HG. The role of numeracy in understanding the benefit of screening mammography. Ann Intern Med 1997; 127: 966-972. Searle J. Routine antenatal screening: not a case of informed choice. Aust N Z J Pub Health 1997; 21: 268-274. Marteau TM, Kidd J, Cook R, et al. Perceived risk not actual risk predicts uptake of amniocentesis. Br J Obstet Gynaecol 1991; 98: 282-286. Slaytor EK, Ward JE. How risks of breast cancer and benefits of screening are communicated to women: analysis of 58 pamphlets. BMJ 1998; 317: 263-264. Col NF, Eckman MH, Wong JB, Pauker SG. Strategies for individualizing patient decisions about hormone therapy. J Clin Endocrinol Metab 1999; 84: 1799-1802. Antman EM, Braunwald E. Acute myocardial infarction. In: Fauci AS, Braunwald E, Isselbacher KJ, et al, editors. Harrison's principles of internal medicine. 14th ed. New York: McGraw-Hill, 1998; 1352-1365. Loop FD. The surgical treatment of artherosclerotic coronary heart disease. In: Schlant RC, Alexander RW, editors. Hurst's The heart. 8th ed. New York: McGraw-Hill, 1994; 1367-1380. Grodstein F, Stampfer MJ, Manson JE, et al. Postmenopausal estrogen and progestin use and the risk of cardiovascular disease. N Engl J Med 1996; 335: 453-461. Falkeborn M, Persson I, Adami H-O, et al. The risk of acute myocardial infarction after oestrogen and oestrogen-progestogen replacement. Br J Obstet Gynaecol 1992; 99: 821-828. Colditz GA, Markinson SE, Hunter DJ, Willett WC. The use of oestrogens and progestins and the risk of breast cancer in post-menopausal women. N Engl J Med 1995; 332: 1589-1593. Gruppo Italiano per lo Studio della Streptochinasi nell'Infarto Miocardico (GISSI). Effectiveness of intravenous thrombolytic treatment in acute myocardial infarction. Lancet 1986; i: 397-402. The Veterans Administration Coronary Artery Bypass Surgery Cooperative Study Group. Eleven-year survival in the Veterans Administration randomised trial of coronary bypass surgery for stable angina. N Engl J Med 1984; 311: 1333-1339. Forrow L, Taylor WC, Arnold RM. Absolute relative: how research results are summarized can affect treatment decisions. Am J Med 1992; 92: 121-123. Malenka DJ, Baron JA, Johansen S, et al. The framing effect of relative and absolute risk. J Gen Intern Med 1993; 8: 543-548. Skolbekken J-A. Communicating the risk reduction achieved by cholesterol reducing drugs. BMJ 1998; 316: 1956-1958. Bunker JP, Houghton J, Baum M. Putting the risk of breast cancer in perspective. BMJ 1998; 317: 1307-1309. SPSS [computer program]. Version 6.1. Chicago Ill: SPSS Inc, 1993. Campbell MJ, Gardner MJ. Calculating confidence intervals for some non-parametric analyses. BMJ 1988; 296: 1454-1456. Calman KC. Cancer: science and society and the communication of risk. BMJ 1996; 313: 799-802. Ayanian JZ, Landrum MB, Normand ST, et al. Rating the appropriateness of coronary angiography -- do practicing physicians agree with an expert panel and with each other? N Engl J Med 1998; 338: 1896-1904. Meisel A, Kuczewski M. Legal and ethical myths about informed consent. Arch Intern Med 1996; 156: 2521-2526. Shahar E. A Popperian perspective of the term 'evidence-based medicine'. J Eval Clin Pract 1997; 2: 109-116. (Received 1 Mar, accepted 19 Aug, 1999) Authors' Details North Western Adelaide Health Service, Queen Elizabeth Hospital, Adelaide, SA. Stephen P Fitzgerald, MB BS, FRACP, Visiting Medical Specialist, Endocrinology; George Phillipov, MSc, PhD, Chief Medical Scientist, Endocrinology. Reprints: Dr G Phillipov, Endocrinology, The Queen Elizabeth Hospital, Woodville, SA 5011. gphillipovATtqehsmtp.tqeh.sa.gov.au 1: Demographics of respondents in the three surveysSurvey 1Survey 2Survey 3 Interventions surveyedHRTHRTHRTTHROMTHROMCABSCABSNumber asked to participate1008298Refusals*1553Actual participants857795SettingHospital outpatients clinics April 1997 General practice surgeries May 1997 General practice surgery October 1997Median age in years (range)60 (17-88)47 (17-78)39 (17-77)Number of men34 (41%)23 (31%)40 (42%)Number born in Australia55 (68%)57 (78%)76 (80%)Number reporting heart disease14 (17%)9 (12%)6 (6%)Number reporting heart surgery11 (13%)6 (8%)2 (2%)*As reported by nursing and reception staff. HRT=hormone replacement therapy. THROM=thrombolysis for acute myocardial infarction. CABS=coronary artery bypass surgery. Back to text 2: Hormone replacement therapy: survey questions and main results A: Surveys 1 and 2 Level of evidence:2 III-2.13-15 There was no difference between groups in surveys 1 and 2 in: age (Mann-Whitney U test, P=0.70); sex (Fisher's exact test, P=0.40); being Australian-born (Fisher's exact test, P=0.44). The difference in median scores was -0.5 (95% CI, -2.1 to +1.0), so data from the two surveys were pooled. Question: Imagine you are age 50. You are offered a therapy which over the next ten years should reduce your risk of heart disease from 10% to 5%. Unfortunately, the safety of this treatment (tablets which you must take every day) is not proven. There is a controversial belief that these tablets might cause cancer in about 2% of individuals. In summary, the tablets have a 5% chance of being helpful, a possible 2% chance of doing harm and a 93% chance of doing neither harm nor good.* How do you think you would feel about such treatment if it were offered to you? Analysis exclusions: 104 patients with age >50 years or reported heart disease. Participants: 58 patients; median age, 38.5 years (range, 17-49 years); 20 (35%) men; 47 (86%) born in Australia. Individual scores independent of:sex (Mann-Whitney U test, P=0.41); age (Kendall correlation coefficient, -0.12; P=0.24); being Australian-born (Mann-Whitney U test, P=0.87). Median response:‡ -2.95 (95% CI, -5 to -2.1). Free-text responses: 32 (55%) responses. Main free-text themes:† 17 (53%), aversion to taking any risk; 7 (22%), prefer alternative medicine procedures. Patients who indicated that their attitude was influenced by the risk of cancer as a side-effect, and those who indicated support for alternative medicine approaches, had a median score of -5. B: Survey 3 Question: Imagine you are age 50. You are offered a therapy which over the next 15 years should reduce your risk of heart disease from 20% to 10%. Unfortunately the safety of this treatment (tablets which you must take every day) is not proven. There is a controversial belief that these tablets might cause cancer in about 1% of individuals. In summary, the tablets have a 10% chance of being helpful, a possible 1% chance of doing harm and an 89% chance of doing neither harm nor good. How do you think you would feel about such treatment if it were offered to you? Analysis exclusions: 27 patients with age >50 years or reported heart disease. Participants: 68 patients; median age, 36 years (range, 17-49 years); 26 (38%) men; 56 (82%) born in Australia. Median response:‡ -0.75 (95% CI, -2.3 to 0). Median score significantly higher than for surveys 1 and 2 (Mann-Whitney U test, P<0.01) *"No effect" statement removed for survey 2. †See Methods for details of the theme categories. ‡Responses ranged from -5 (definitely would not accept the treatment) to +5 (definitely would accept the treatment). Back to text 3: Thrombolysis for acute myocardial infarction, and coronary artery bypass surgery: survey questions and main results Responses for surveys 1 and 2 were combined. Analysis exclusions: 51 patients with age >65 years. Participants: 111 patients; median age, 47 years (range, 17-64 years); 37 (34%) men; 79 (75%) born in Australia; 21 (19%) reported cardiovascular disease. Individual scores independent of: sex (Mann-Whitney U test, P=0.07); age (Kendall correlation coefficient, P=0.19); being Australian-born (Mann-Whitney U test, P=0.70); prior heart surgery (Mann-Whitney U test, P=0.61). Thrombolysis for acute myocardial infarction Level of evidence:2 II.16 Question: Imagine you are 65. You are brought to hospital with severe chest pain and it is thought that you are having a heart attack. You are offered an injection which should reduce your chance of dying from 10% to 7%, but unfortunately has a side-effect of causing serious bleeding in about 1% of patients. If, for example, this bleeding occurs in the brain, it might cause a stroke or death. In summary there is a 3% chance of the injection saving your life, a 1% chance of causing serious bleeding and a 96% chance that it will do neither*. How would you feel about such an injection? Median response:‡ -0.5 (95% CI, -0.9 to 0) Free-text response: 55 (50%) responses. Main free-text themes:† 14 (25%), odds of success given; 12 (22%), aversion to taking any risk; 11 (20%) willing to take any risk for health. Coronary artery bypass surgery Level of evidence:2 II.17 Question: Imagine in the above scenario that you survived the heart attack which fortunately was small and that you now feel perfectly well. As part of follow-up tests it is discovered that you have blockages in your coronary arteries (blood vessels to the heart). It is possible to unblock these arteries with major surgery. This surgery has about 1 in 10 chance of saving your life sometime over the next 5 years. Unfortunately as well as having to endure major surgery, there is a 1 in 20 chance that the surgery will cause death or a stroke. In summary there is a 10% chance that the surgery will be helpful, a 5% chance that it will cause harm and an 85% chance that it will make no difference*. How would you feel about an operation? Median response:‡ -0.1 (95% CI, -0.5 to +1.3) Free-text response: 46 (41%) responses. Main free-text themes:† 10 (22%), odds of success given; 9 (20%), aversion to taking any risk. Patient attitude to THROM and CABS was less negative than for HRT (Wilcoxon matched-pairs signed-ranks test, P<0.001). Difference in median scores between patients with and without heart disease was significantly increased for CABS (+2.0; 95% CI, 0 to +3.8), but not for THROM (+1.2; 95% CI, -0.6 to +2.8). *"No effect" statement removed for survey 2. †See Methods for details of the theme categories. ‡Responses ranged from -5 (definitely would not accept the treatment) to +5 (definitely would accept the treatment). CABS=coronary artery bypass surgery. HRT=hormone replacement therapy. THROM=thrombolysis for acute myocardial infarction. Back to text

Stephen P Fitzgerald · George Phillipov

Notable Cases Fulminant hepatitis A in Indigenous children in north Queensland Jeffrey N Hanna, Tim H Warnock, Ross W Shepherd and Linda A Selvey Since 1993, three Indigenous children in north Queensland have died of fulminant hepatitis A. Even if the children had been able to undergo liver transplantation, prolonged immunosuppressant therapy and the likelihood of opportunistic infections would inevitably have jeopardised any chance of long-term survival. As hepatitis A has become a leading infectious cause of death in young Indigenous children in north Queensland, hepatitis A vaccine has recently been introduced into the vaccination schedule for these children. MJA 2000; 172: 19-21 For editorial comment, see McCaughan & Torzillo Introduction - Clinical record 1 - Clinical record 2 - Clinical record 3 - Discussion - Acknowledgement - References - Authors' details - - More articles on Aboriginal health Introduction Fulminant hepatitis A is rare in children, and most recent reports have been from developing countries.1-4 Occasional cases in children from industrialised countries tend to be in those who have recently travelled to hepatitis A endemic areas5,6 or who live in conditions of considerable socioeconomic disadvantage.7We describe three cases of fulminant hepatitis A in Indigenous children in north Queensland, all of whom died. This uncommon but serious manifestation of a relatively common infection in Indigenous children presents major difficulties in clinical management,8 illustrating the importance of preventive public health measures. Clinical record 1 A 4.5-year-old Torres Strait Islander boy, who had previously been in good health, presented to a local health centre in early 1993 with anorexia, dark urine and jaundice of several days' duration. Vomiting and irritability persisted for five days, and he was evacuated to Thursday Island Hospital. Two days later, he had a generalised seizure and became semi-comatose after an episode of haematemesis, and was evacuated to Cairns Base Hospital. On admission in Cairns, the boy was deeply jaundiced, with hepatic fetor and moderate hepatomegaly. He did not respond to painful stimuli, and was generally hypertonic with upgoing plantar reflexes. He had raised serum bilirubin and liver enzyme levels, markedly elevated serum ammonia level, hypoglycaemia and a bleeding diathesis (see Box). He was positive for hepatitis A IgM, confirming acute hepatitis A; subsequent serological tests were negative for hepatitis B and C, cytomegalovirus (CMV) and herpes simplex virus (HSV) infection. The boy was stabilised, placed on ventilator support and transferred the next day to the Royal Children's Hospital in Brisbane for consideration of liver transplantation. However, on arrival in Brisbane, he had fixed, dilated pupils, and urgent computed tomography showed gross cerebral oedema. Despite intensive treatment, his condition deteriorated, and he died the next day, nine days after initial presentation. Postmortem examination showed massive hepatic necrosis and diffuse hypoxic ischaemic brain injury, secondary to cerebral oedema and raised intracranial pressure. Clinical record 2 A 4.5-year-old Aboriginal girl was admitted to a community hospital near Townsville in late 1997 with fever, lethargy, dark urine and jaundice of uncertain duration. Apart from a persistent tinea capitis infection, she had been in reasonably good health. Four months before the current illness, she had been treated with a four-week course of ketoconazole, and results of liver function tests were normal at that time. On admission, she had raised serum bilirubin and liver enzyme levels (Box). Serological tests confirmed acute hepatitis A; subsequent tests were negative for hepatitis B and C, CMV, HSV and Epstein-Barr virus infection. Four days after her admission, the girl's family took her from the hospital, and on her return the next day she was delirious. She was transferred to Townsville General Hospital. On admission in Townsville, she was deeply jaundiced and irritable, with a depressed level of consciousness and moderate hepatomegaly. Peripheral reflexes were brisk, with up-going plantar reflexes and bilateral ankle clonus. Her serum bilirubin level had risen further (Box), she was hypoglycaemic, had a bleeding diathesis, and soon became hypokalaemic (serum potassium level, 2.7 mmol/L [RR, 3.5-4.5 mmol/L]). Several hours after admission, she had an episode of profound hypotension (systolic pressure, 50 mmHg) after aspiration of a large volume of bloodstained fluid from the nasogastric tube. She was transferred the next day to the Royal Children's Hospital in Brisbane for consideration of liver transplantation. On arrival in Brisbane, her serum ammonia and sodium levels were both markedly elevated (ammonia, 158 µmol/L; sodium, 151 mmol/L [RR, 133-143 mmol/L]). Computed tomography showed cerebral oedema with a possible ischaemic lesion in the right cerebral hemisphere. She was listed for urgent liver transplantation pending a suitable donor, but the next day developed signs of raised intracranial pressure which did not respond to medical management. She died eight days after initial presentation. Clinical record 3 A 2.5-year-old Aboriginal girl, who had previously been in good health, presented to a health centre in a Cape York community in September 1998 with malaise and jaundice of one day's duration. She was reviewed the next day by the visiting Royal Flying Doctor Service; a clinical diagnosis of acute hepatitis was made, but, because she did not appear particularly unwell, no specific treatment was requested. Liver function tests on that day (Day 2) showed her serum concentrations of bilirubin, AST and ALT were raised (Box). Serological testing confirmed acute hepatitis A, and tests for hepatitis B and C were negative. The girl was reviewed two days later by health centre staff. She reportedly had stopped eating and appeared lethargic, although she responded appropriately to voices and other stimuli. When seen the next morning, she had deteriorated markedly, no longer responded to her environment and was obtunded and dehydrated, with grunting respirations. She was evacuated urgently to Cairns Base Hospital. On admission in Cairns she was deeply jaundiced, with hepatic fetor and mild hepatomegaly. She was semi-comatose, had brisk symmetrical reflexes and upgoing plantar reflexes. Her serum ammonia level was markedly raised and she had a bleeding diathesis (Box). Urgent transfer to Brisbane was arranged, but she deteriorated nine hours after admission with extensor posturing and diminished respiratory effort; her limbs subsequently became flaccid and areflexic. She died several hours later, five days after initial presentation. Discussion All three of the reported children were critically ill with fulminant liver failure and grade III hepatic encephalopathy9 when referred to specialist services in Cairns or Townsville. Mortality rates in fulminant hepatitis approach 70% in children,8 with death usually occurring within 8-10 days. Youth and severity of hepatic encephalopathy are both strong predictors of poor outcome;10 the main cause of death is cerebral oedema.10Liver transplantation is the only option for children with fulminant hepatitis A and advanced stages of encephalopathy who are considered unlikely to survive with medical supportive therapy alone.6,8-10 Published criteria for transplantation include either PT > 100 s or any three of the following: PT > 50 s, age < 10 years or > 40 years, jaundice for > 7 days before onset of encephalopathy, or serum bilirubin > 300 µmol/L.11 However, in practice, the availability of a suitable donor organ may be the critical issue, and therefore most transplant centres list patients as soon as transplantation is considered, and make a final decision if and when a donor organ becomes available.8 Two of the children (Patients 1 and 2) clearly met the above criteria and were listed for urgent transplantation on arrival in Brisbane, but it soon became apparent that irreversible brain damage had occurred. However, even if they had received a successful liver transplant, the difficulties of monitoring prolonged immunosuppressant therapy and the likelihood of opportunistic infections in their home communities would have inevitably jeopardised any chance of long-term survival. Hepatitis A has been the most frequent infectious cause of death in preschool-aged Indigenous children in north Queensland since 1993 (excluding neonatal deaths and pneumococcal infections, for which data are incomplete) (J Hanna, unpublished data). Two deaths resulted from the "standard" vaccine-preventable diseases (measles and Haemophilus influenzae type b infection), and one from meningococcal infection. Over the same period, no non-Indigenous children died of hepatitis A in north Queensland. We can only speculate that "host factors", as well as the quantity of viral inoculum,6 may determine why some Indigenous children develop fulminant hepatitis A. The notification rates of hepatitis A in children under five years of age in north Queensland in 1996-1997 were 264 and 10 per 100 000 in Indigenous and non-Indigenous children, respectively.12 About 25% of notified cases of hepatitis A in Indigenous people in north Queensland occur in children under five years of age,12 indicating considerable circulation of the hepatitis A virus among these young children. However, nearly half the notified cases in Indigenous people in north Queensland occur in children aged 5-14 years,12 indicating that many Indigenous primary school-aged children are susceptible to the infection. Very few notifications are in Indigenous adults over 30 years of age. Because hepatitis A is common in Indigenous children in north Queensland, and because of the demonstrated unsatisfactory outcome in those few children who develop fulminant liver failure, the emphasis must be on preventing infection. Hepatitis A virus is readily transmitted in environments with inadequate sanitation and water supply, suboptimal hygiene and overcrowding.13 The current status of housing and environmental health infrastructure throughout Indigenous Australia has recently been detailed elsewhere, and others have described some of the health consequences of these circumstances.14,15 Despite considerable recent improvements in provision of adequate housing and sanitation infrastructure in several Indigenous communities in north Queensland, many communities still have unresolved problems of overcrowding, water supply and waste disposal.16 Some communities continue to experience environmental contamination by raw sewage during wet season floods, and several Torres Strait Island communities experience severe water shortages in the dry season.17 Some of these problems are likely to persist for the foreseeable future. Paradoxically, as environmental circumstances improve, the age of reported patients will likely shift upwards, and the "visibility" of hepatitis A -- and public health concern -- will increase.13 Indigenous communities in North America have experienced cyclical outbreaks of hepatitis A every 5-10 years for many years,18 a pattern that may become more obvious in Indigenous Australian communities. Inactivated vaccines with proven effectiveness in preventing hepatitis A in children19,20 have recently become available in Australia. These vaccines have also been shown to interrupt transmission in "closed" communities prone to recurrent outbreaks,21,22 leading to the recommendation of routine hepatitis A vaccination for Indigenous children in North America.18 An inactivated hepatitis A vaccine has been offered to Indigenous children in north Queensland since early 1999. Two doses six months apart have been recommended, integrated as far as possible into the standard vaccination schedule, with the first dose at 18 months of age and the second at 24 months, necessitating only one "extra" visit to a vaccination service. Current resources allow for catch-up hepatitis A vaccination up to the sixth birthday; about 5500 Indigenous children aged 2-5 years live in north Queensland. Young children play a particularly important role in maintaining transmission of hepatitis A virus to older susceptible individuals during community-wide outbreaks,24 because of their lack of bowel control and attention to hygiene, and need for adult supervision of their toileting needs. Therefore, vaccination of preschool-aged Indigenous children provides a means not only to prevent further cases of fulminant hepatitis A in Indigenous Australian children, but also to reduce the extent of future outbreaks in their communities. A particular challenge will be to ensure that Indigenous children in rural towns and urban settings gain access to the vaccine.25 Acknowledgement We wish to thank the families of the three children for allowing us to describe their children's final illness. References Zacarias J, Brinck P, Cordero J, Velasco M. Etiologies of fulminant hepatitis in pediatric patients in Santiago, Chile. Pediatr Infect Dis J 1987; 6: 686-687. Ozsoylu S, Kocak N. Acute hepatic failure related to hepatitis A [letter]. Lancet 1989; 313: 901. Ciocca M, Ramonet M, Cuarterolo M, et al. Fulminant hepatic failure for viral hepatitis in children. IX Triennial International Symposium on Viral Hepatitis and Liver Disease; Rome; 1996 21-25 Apr. Abstract no. A 316. Arora NK, Nanda SK, Gulati S, et al. Acute viral hepatitis types E, A and B singly and in combination in acute liver failure in children in north India. J Med Virol 1996; 48: 215-221. Masada CT, Shaw BW, Zetterman RK, et al. Fulminant hepatic failure with massive necrosis as a result of hepatitis A infection. J Clin Gastroenterol 1993; 17: 158-162. Debray D, Cullufi P, Devictor D, et al. Liver failure in children with hepatitis A. Hepatology 1997; 26: 1018-1022. Friedland IR, Zuckerman M, Kala UK, Parbhoo KB. Fulminant hepatitis in children: report of 12 cases. Ann Trop Paediatr 1991; 11: 207-211. Shepherd RW. The treatment of end-stage liver disease in childhood. Aust Paediatr J 1988; 24: 213-216. Suchy FJ. Fulminant hepatic failure. In: Behrman RE, Kliegman RM, Arvin AM, Nelson WE, editors. Nelson textbook of pediatrics. 15th ed. Philadelphia: WB Saunders, 1996: 1150-1152. Hoofnagle JH, Carithers RL, Shapiro C, Ascher N. Fulminant hepatic failure: summary of a workshop. Hepatology 1995; 21: 240-252. O'Grady J. Management of acute and fulminant hepatitis A. Vaccine 1992; 10 Suppl 1: S21-S23. Merritt T, Symons D, Griffiths M. The epidemiology of acute hepatitis A in north Queensland, 1996-1997. Commun Dis Intell 1999; 23: 120-124. Gust ID. Epidemiological patterns of hepatitis A in different parts of the world. Vaccine 1992; 10 Suppl 1: S56-S58. McLennan W, Madden R. The health and welfare of Australia's Aboriginal and Torres Strait Islander peoples. Canberra: AGPS, 1997: 11-19. (ABS Catalogue no. 4704.0.) Mathews JD. Historical, social and biological understanding is needed to improve Aboriginal health. Recent Adv Microbiol 1997; 5: 257-334. Legislative Assembly of Queensland: Public Works Committee. The provision of infrastructure in Cape York. Report no. 38, June 1997. Henderson G, McKenna P, Kingsley A, et al. Hepatitis A and water supply in the Torres Strait Area. Aboriginal Torres Strait Islander Health Inform Bull 1995; 21: 48-58. Welty TK, Darling K, Dye S, et al. Guidelines for prevention and control of hepatitis A in American Indian and Alaska Native communities. S D J Med 1996; 49: 317-322. Werzberger A, Mensch B, Kuter B, et al. A controlled trial of formalin-inactivated hepatitis A vaccine in healthy children. N Engl J Med 1992; 327: 453-457. Innis BL, Snitbhan R, Kunasol P, et al. Protection against hepatitis A by an inactivated vaccine. JAMA 1994; 271: 1328-1334. Werzberger A, Kuter B, Nalin D. Six years' follow-up after hepatitis A vaccination [letter]. N Engl J Med 1998; 338: 1160. McMahon BJ, Beller M, Williams J, et al. A program to control an outbreak of hepatitis A in Alaska by using an inactivated hepatitis A vaccine. Arch Pediatr Adolesc Med 1996; 150: 733-739. Australian Bureau of Statistics. 1996 Census of population and housing. Smith PF, Grabau JC, Werzberger A, et al. The role of young children in a community-wide outbreak of hepatitis A. Epidemiol Infect 1997; 118: 243-252. Hanna JN, Malcolm RL, Vlack SA, Andrews DE. The vaccination status of Aboriginal and Torres Strait island children in Far North Queensland. Aust N Z J Public Health 1998; 22: 664-668. (Received 6 Apr, accepted 24 Aug, 1999) Authors' details Queensland Health, Cairns, QLD. Jeffrey N Hanna, MPH, FAFPHM, Public Health Physician, Tropical Public Health Unit; Tim H Warnock, FRACP, Paediatrician, Cairns Base Hospital. Department of Gastroenterology and Hepatology, Royal Children's Hospital, Brisbane, QLD. Ross W Shepherd, MD, FRACP, Director. Communicable Diseases Unit, Queensland Health, Brisbane, QLD. Linda A Selvey, PhD, FAFPHM, Manager. Reprints will not be available from the authors. Correspondence: Dr J N Hanna, Tropical Public Health Unit, Queensland Health, PO Box 1103, Cairns, QLD 4870. md1AThealth.qld.gov.au Results of investigations of three Indigenous children with fulminant hepatitis AReferencePatient 2Patient 3TestrangePatient 1On admissionDay 5Day 2Day 5Serum bilirubin level (µmol/L)<20389166269225202Aspartate aminotransferase level (U/L)10-6016016785137853672279Alanine aminotransferase level (U/L)<4017502690187322851352Blood glucose level (mmol/L)3.5-6.01.7ND1.9ND3.2Prothrombin time (s)10-15>180ND147ND84Activated partial thromboplastin time (s)23-37135ND107ND65Serum ammonia level (µmol/L)10-35416NDNDND205ND = not done

Jeffrey N Hanna · Tim H Warnock · Ross W Shepherd · Linda A Selvey

For Debate An integrated electronic health record and information system for Australia? Christopher D Mount, Christopher W Kelman, Leonard R Smith and Robert M Douglas An integrated health record and information system, although costly and difficult to implement, would provide benefits for clinicians and patients through better clinical care, and for the healthcare system through better data for policy development and resource allocation. MJA 2000; 172: 25-27 Introduction - Health records - The Integrated Health Record and Information System - Benefits - Implementation - Conclusion - Acknowledgements - References - Authors' details - - More articles on Informatics and computers

Christopher D Mount · Christopher W Kelman · Leonard R Smith · Robert M Douglas

For Debate Integrated electronic health records and patient privacy: possible benefits but real dangers Strategies to develop integrated electronic health records must address consumer concerns, not dismiss them with claims of the "public good" Meredith Carter MJA 2000; 172: 28-30 Introduction - Newspeak - Global claims - New uses of data - Attention to consumer and community - References - Authors' details - - More articles on Informatics and computers

Meredith Carter