Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Public health aspects of genetic screening for hereditary haemochromatosis in Australia

Public health aspects of genetic screening for hereditary haemochromatosis in Australia Abstract Hereditary haemochromatosis (HH) is an inherited disorder of iron absorption. It meets several of the key public health principles for population-based screening and is considered to be a test-case for public health genetics. However, there has been relatively little debate in the public health or wider community regarding the merits of population-based genetic screening for HH. Genetic susceptibility to HH occurs in about 1:200 people and although mortality is low (age-standardised rate 2.75/million), there are potentially serious clinical manifestations of iron overload. Regular venesection is a simple and effective treatment for early stage iron overload. DNA-based testing is available and iron overload may be identified using serum transferrin saturation and ferritin tests. However, there are impor tant gaps in knowledge relevant to screening for HH. The limited data on penetrance of HFE genotypes, and thus the uncer tain probability that genetically susceptible individuals will develop clinically significant disease, is a major impediment to population-based genetic screening. Clinical evidence supports treating ear lystage disease but no randomised controlled trials of the effectiveness of screening in reducing the burden of disease have been conducted. In addition, the natural history of early stages of HH and factors that may modify progression are unclear. Two international consensus panels on HH concluded that there is insufficient evidence for population-based screening at present. We present recommendations to advance the debate on screening for HH in Australia. (Aust N Z J Public Health 2002; 26: 518-24) Dorota M. Gertig, Ashley Fletcher and John L. Hopper Centre for Genetic Epidemiology School of Population Health, , University of Melbour ne, Victoria ereditary haemochromatosis (HH) is a disorder of iron overload that may occur in individuals who have an inherited susceptibility due to mutations in the HFE gene. Increased absor ption of dietary iron leads to a potentially har mful accumulation of iron in parenchymal cells of various organs and tissues, in particular the liver, heart and pancreas. Two common mutations (C282Y and H63D) account for more than 90% of all cases of HH. The onset of disease is age-dependent and usually manifests in middle age with non-specif ic symptoms such as f atigue, arthralgia, impotence and arthritis. As iron overload increases, symptoms may become more severe and in later life it may lead to severe disease that includes liver cirrhosis, cardiomyopathy and diabetes.1 The classic picture of HH, known as ‘bronzed diabetes’ is a rare condition that results from severe or gan damage due to iron deposition. Since the identification of the HFE gene mutations responsible for HH in 1996,2 the prospect of population-based genetic screening for this condition has been debated. HH fulf ils many of the public health criteria for diseases suitable for screening. In fact, it has been proposed that screening for HH may become a model for future population-based genetic screening of other hereditar y conditions as appropriate DNA-based tests and therapeutic inter ventions become availab le.3 Therefore, it is par ticularly important to ensure that the public health implications of a screening program for HH are thoroughly considered. There has been substantial debate in the literature regarding the optimal screening approach for HH.4 Genetic screening refers to testing individuals directly for genetic susceptibility using a DNA-based test. Although this approach will detect a proportion of people at an earlier stage of disease prior to development of iron overload, it is unclear how many genetically susceptible people will go on to develop clinically signif icant disease. Phenotypic screening refers to identifying individuals with iron overload initially using transfer rin saturation (TS) and serum fer ritin (SF) tests. The sensitivity and specificity depend on the cut-off used and phenotypic screening will detect people who have iron overload and def iciency from other causes.5 However, the natural histor y of iron overload is uncertain and issues such as laboratory standardisation, education of medical practitioners and cost-effectiveness still need to be resolved. Another approach is case-f inding through testing the relatives of people with HH or people with symptoms or conditions that may be related to iron overload, such as f atigue, diabetes, etc. While screening of relatives is widely recommended, there are at present few data on the predictive value of various symptoms related to iron overload in detecting HH. Because HH is considered to be a test-case for public health genetics, in this review we concentrate on public health aspects of genetic screening for HH. Public health criteria Screening programs for all diseases, both hereditary and non-hereditary, should meet Submitted: May 2002 Revision requested: July 2002 Accepted: September 2002 Correspondence to: Dorota Gertig, Centre for Genetic Epidemiology, School of Population Health, University of Melbourne, 723 Swanston Street, Carlton, Victor ia 3053. Fax: (03) 9347 6929; e-mail: d.gertig@unimelb.edu.au AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH VOL. 26 NO. 6 Policy Public health aspects of genetic screening Table 1: Public health criteria for screening. Public health criteria The condition The condition should be an important health problem. The natural histor y of the disease should be understood. Hereditary haemochromatosis should be recognisable at an early stage. The treatment Treatment at an early stage should be more effective than treatment at a late stage. The test There should be a suitable test, in ter ms of sensitivity, specificity and positive predictive value, and the test should be acceptable to the population. For diseases of insidious onset, screening should be repeated at intervals deter mined by natural histor y. The screening program Physical and psychological harm to those screened should be less than the benefit. There should be adequate facilities for diagnosis and treatment. Does genetic screening for HH meet these criteria? Yes. History of early stage disease not well understood. Penetrance uncer tain. Iron overload detectable ear ly but case definition complicated by uncertain natural histor y and non-specific clinical picture. Yes, based on Level III and IV evidence. Yes. Laboratory standardisation of transferrin saturation (TS) and serum ferritin (SF) an issue. No need to repeat genetic screening test but TS and SF must be monitored in C282Y homozygotes. Insufficient evidence. No RCTs of screening have been conducted. Yes, potentially, but depends on the extent of screening and treatment. defined criteria to assess suitability and appropriateness, in particular that: • screening must result in more benef it to the targeted population than harm; and • early treatment, initiated because of screening, is shown to be effective in reducing morbidity or mortality from the disease.6 The public health issues for genetic screening (see Table 1) are similar to screening for non-hereditary diseases, although the terminology differs and there are some special considerations particularly with respect to implications for f amily members. First, genetic screening does not identify individuals who actually have disease; rather, the presence of a genetic prof ile associated with a particular disease only indicates the individual is at increased risk. This is similar to screening for elevated cholesterol as a risk factor for heart disease. The probability that an individual with a particular genetic profile will develop an outcome of interest (e.g. a specif ic disease or set of symptoms) by a cer tain age (cumulative age-specif ic risk) is referred to as the ‘penetrance’ of the genotype for that outcome. For most hereditary disorders the penetrance is ‘incomplete’, that is the chance of developing a disease is not 100% and it may be modif ied by environmental and other genetic factors. For example, different mutations in the BRCA1 and BRCA2 genes are associated with different lifetime risks of breast, ovarian and other cancers. The penetrance of these mutations appears to vary by the position of the mutations within the gene and by other genetic and environmental influences.7 Second, the potential adverse effects of ‘labelling’ a person with a genetic susceptibility when they have not yet developed or may never develop disease, must be addressed. These include psychological effects and the potential for discrimination. Third, genetic screening has consequences for family members. If the decision is made to inform family members of individuals found to have genetic risk, the provision of appropriate counselling and other issues must be considered. 2002 VOL. 26 NO . 6 Is HH an important health problem? Table 2 shows that the C282Y and H63D mutations are rare in Aboriginal and Asian populations and that the C282Y frequency is higher in northern than southern European populations. The frequency of these mutations in Australia and New Zealand is most comparable to nor thern European populations so that the C282Y allele frequency translates to between 1:150 and 1:300 being homozygotes. The estimated proportion of HH due to C282Y homozygosity (two copies of the C282Y mutation, one from each parent) ranges from 33% in southern Italy, 69% in northern Italy, 8 and up to 100% in Australia and the United Kingdom.9,10 Because homozygosity for the H63D mutation appears to account for a much smaller proportion of HH than the C282Y mutation,11 screening for that mutation has not generally been advocated. Haemochromatosis due to causes other than HFE mutations is rare; therefore the relative risk of HH due to HFE mutations is high. The odds of dev eloping iron ov erload for C282Y homozygotes compared with the odds of developing iron overload in controls are very high (estimates of odds ratio (OR) ~2,000 and 4,000 respectively). 11 Reported risks for other genotypes are much smaller: OR=25 and 32 for genotype C282Y/H63D; OR=8.6 and 5.7 for H63D/H63D. 11,12 The validity of these estimates is uncertain due to the v aried inclusion criteria for cases, the range of case def initions and a lack of comparable control groups in several of the studies. In 1998, the age-standardised rate of haemochromatosis-related deaths in Australia was 2.75 per million, compared with 1.8 per million in the US in 1992.13 Mor tality is likely to be underestimated due to the variety of clinical manifestations from the disease and under-diagnosis of the condition. Is the natural history of HH understood? The natural history of HH has been studied predominantly AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Gertig, Fletcher and Hopper Article Table 2: HFE genotype and allele frequencies from Australian and New Zealand studies, with summaries of other regions. Year Australia 199738 1998 Size Study design Source C282Y C282Y 0.5% 0% 0.5% 0% C282Y – 11.2% 1.1% 11.9% 13.0% H63D H63D H63D – C282Y H63D C282Y (%) H63D (%) 1,660 Cross-sectional 185 3,011 69 Cross-sectional Prospective Casecontrol Newbor ns Australian Aborigines General population Liver disease (not HH) General population 0.5% 2.1% 4.3% 3.8% 23.6% 20.3% 2.2% 1.4% 199914,60 199940 New Zealand 199741 199842 1,032 Cross-sectional 1.0% 0.5% 0.7% 0.02% 18.6% 11.4% 12.6% 3.4% 2.3% 2.2% 2.6% 22.6% 23.2% 23.3% 1.8% 2.4% 0.7% 10.3 7.0 (Range) 8.2 (6.2-9.9) 1.6 (0.2-3.7) 6.2 (5.3-6.6) 0.1 (0.0-0.2) 14.4 (Range) 15.0 (11.5-15.3) 12.6 (11.9-20.6) 15.1 (14.6-15.4) 0.4 1,064 Cross-sectional Electoral roll Europe (Summary) Northern 43-48 12,733 Southern 8,49-53 4,598 US (Summary) Caucasians17,54-56 12,534 Asia (Summary) Asians57-59 0.5% 9.3% 2.4% 23.2% 2.1% 1,350 0% 2.0% 0% 2.0% 0% among people with advanced disease. Cases identif ied because a relative has HH are more likely to be symptomatic due to the sharing of other genetic or environmental modifying factors.14 There are a few retrospective studies of people with HH, only some of which included individuals who were untreated,15 but these are not generalisable to population screening as the majority of people with HH had cir rhosis at presentation. The only pub lished prospective data of the natural history of iron overload come from the Busselton study in Western Australia 14 that included 12 untreated C282Y homozygotes (mean age 49, range 30 to 74), all of whom had elevated TS and more than half (58%) had elevated SF at the end of four years of follow up. HFE penetrance Estimates of penetrance of HFE genotypes depend on age, sex and the outcomes used to define penetrance, i.e. iron overload or various clinical disease endpoints. There are relatively few reliable estimates of penetrance of the C282Y homozygous genotype and, homozygotes had clinical features consistent with iron overload; however, the frequency of symptoms was not compared with a control series. In a primary-care population, Beutler and colleagues also reported a higher average iron index in homozygotes (n=152)17 but found the prevalence of clinical symptoms among C282Y homozygotes to be less than 1%.18 Thus the penetrance of C282Y homozygosity for iron overload may be approximately 50%, but the penetrance for clinical disease may be ver y low. Modifiers of HFE penetrance Genetic and environmental factors (including dietar y factors) that are known or likely to affect iron absorption or excretion may be potential modif iers of disease expression in genetically susceptible individuals.3 Both age and gender are important in general, these estimates are not age-specif ic. In a pooled analysis of case-control studies the estimated penetrance for iron overload in C282Y homozygotes was between 31 and 94%, depending on the prevalence of iron overload due to disease and the prevalence of C282Y homozygosity.11 An earlier pooled analysis of seven population-based studies16 estimated penetrance to be 50% for males and 44% for females. The Busselton study in Wester n Australia found penetrance of C282Y homozygosity for raised serum fer ritin was 58%14 based on 12 homozygotes. Four of these determinants of body iron stores. In the general population, iron levels increase with age and are lower in women due to loss of iron through regular menstruation and the effects of pregnancy.19 Dietary f actors, such as red meat intake and alcohol intake, have been shown to be predictors of serum fer ritin, but to different degrees for men and women.20,21 A recent study of Australian twin pairs found that after adjustment for age, gender and body mass index, HFE genotype explained only a small proportion of the variation in iron stores and about half the remaining variance in serum ferritin was accounted for by genetic f actors other than mutations in HFE.22 Animal models suggest that genes in the iron metabolism pathway may play a role in modifying HFE phenotype.23 Additional data will be forthcoming from a large study funded by the National Institutes of Health (NIH) in the US that VOL. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 26 NO. 6 Policy Public health aspects of genetic screening Figure 1: Age-specific haemochromatosis-associated mortality in Australia, by sex, 1997/98. 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ Ma les Fe ma les Data were obtained on the cause of death as presented on death cer tificates from the Austr alian Institute of Health and Welfare to estimate the mortality from HH in Austr alia. As there is no specific ICD9 code for haemochromatosis, the search was performed in accordance with previously established methods;13 briefly, iron disorders with known codes were subtracted from the general code 275.0 (disorders of iron metabolism, which includes haemochromatosis). There w ere 141 haemochromatosis-associated deaths (114 males and 27 females) in 1997 and 1998. Mortality Rate per Million Age is enrolling more than 100,000 people in primary care settings. This study will evaluate penetrance of HFE mutations, modifiers of penetrance and acceptability of screening. Is there a suitable test for HH and is the test acceptable to the population? A suitable screening test must be accurate and reliable. There are two screening tests for HH: a DNA-based test that identif ies people with genetic susceptibility and a phenotypic test that def ines iron overload. In 1998, testing for the C282Y mutation of the HFE gene became the f irst genetic test eligible for a Medicare Australia rebate.29 The test is reported to be highly accurate in detecting the C282Y and H63D mutations b ut sensitivity for predicting HH depends on the propor tion of HH due to the common HFE mutations in the screened population. For example, for southern Europeans this propor tion may be as low as 60%. Phenotypic testing involves assessment of TS and SF levels, which are relativel y inexpensive and ha ve reasonab le sensitivity and specificity, although elevated levels need to be followed up with a repeat fasting sample. Is HH recognisable at an early stage? The case definition for HH includes the presence of a deleterious HFE genotype and iron overload (as defined by raised TS and SF). Clinical signs and symptoms of conditions considered to be caused by iron overload may be present but are not necessarily part of the def inition and may be non-specif ic. Although early stages of iron overload are readily detectable using TS and SF, these levels are dependent on age and fasting status. In the Australian context approximately 10% of people with HH will not carry HFE mutations.24 Is treatment at an early stage more effective than treatment at a late stage? The standard treatment of early stage HH (i.e. before serious liver damage) is periodic phlebotomy. Treatment is usually lifelong, with on average four venesections per year needed to maintain normal iron levels. Annual assessment of TS and SF levels is recommended.25 At present, support for the effectiveness of early treatment is based largely on Level III and IV evidence (limited cohort and time series studies, and expert opinion, respectively). There have been no randomised controlled trials of the effectiveness of treatment for early stage HH. For patients with advanced HH, survival at five and 10 years is better than for untreated patients, after adjustment for age and clinical conditions.15,26 Studies have shown that survival in patients with HH who have undergone regular treatment, and who did not have cirrhosis or diabetes at presentation, is comparable to that of the general population.27,28 Therefore, the weight of clinical evidence suggests that earlier diagnosis and treatment of advanced HH improves survi val by several years and that individuals with HH who are treated at an early stage, i.e. do not have severe organ damage, will most likely have a normal life expectancy. Screening should be repeated at intervals determined by natural history As genetic testing is highly accurate and genetic susceptibility remains constant over a person’s lifetime, it should only be necessar y to monitor serum iron markers of individuals who are C282Y homozygotes (and possibly compound heterozygotes) at regular inter vals. The frequency of such monitoring will vary according to the rate at which de-ironing occurs. Physical and psychological harm should be less than the benefit The most important criterion for a screening program is that it results in sufficient benef it to offset any physical or psychological har m to the screened population, including their families. There should be evidence from high-quality randomised controlled trials (RCTs) that the screening program is effective in reducing mor tality and morbidity. In contrast to screening for other chronic conditions (e.g. mammographic screening for breast cancer), there have been no RCTs of screening for HH. Conduct of an RCT for HH would require a better understanding of the natural history of 2002 VOL. 26 NO . 6 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Gertig, Fletcher and Hopper Article iron overload, as it is unclear what the appropriate endpoints for such a trial would be. Therefore, at present there is inadequate evidence to assess whether screening results in more benefit to the population than harm. Some of the possible adverse effects of screening are false positives, false negatives, labelling, costs and premature implementation.30 Of par ticular concern, especially with respect to genetic screening, is that the screening program must be socially and ethically acceptable to health professionals and the public. Although false positives for genetic results are rare due to the high accuracy of DNA-based testing for specif ic mutations, the same does not apply to phenotypic screening. False positives may occur after follow-up testing for serum iron mark ers, although consistent false positive results are unlikely over time. False negative results from genetic testing may also occur as described above. Few studies have assessed the issues of labelling, stigmatisation and discrimination with respect to genetic screening for HH. In a small Canadian study it was found that genetic testing for HFE mutations was well accepted.31 On a f amily level there is the question of how other members react to a relative found to be at genetic risk. Issues range from the obligation of whether to tell other members, to others not wanting to be tested due to problems with health insurance. Data on the overall attitude of health insurers to people who are at genetic risk are not available, but there are individual case reports indicating that this may be a problem.32,33 An agreement has been made with the umbrella organisation for life and disability insurance in Australia (Investment and Financial Ser vices Association) and the Haemscreen study regarding life insurance policies for people undergoing screening for HH. This unique agreement will mean that homozygotes without organ damage and w ho have normal iron levels or whose iron levels retur n to normal following treatment will obtain policies at baseline rates.34 Cost effectiveness A number of factors affect the cost-effectiveness of screening, including the cost of genetic testing (which is likely to decrease over time as testing becomes more widespread), the sensitivity would require pre- and post-test genetic counselling as well as regular monitoring of af fected and unaffected C282Y homozygotes and treatment of iron overload in affected individuals. Based on an estimated prevalence of the C282Y genotype in people of northern European origin of 0.5%, 500 homozygotes would be detected for ever y 100,000 people screened. Using current estimates of penetrance, approximately half of the detected homozygotes over their lifetime would require treatment and regular monitoring; the other half would require monitoring of serum iron studies on an ongoing basis. The international perspective Since the characterisation of the HFE gene in 1996, there have been two major international consensus conferences examining the suitability of HH for population-based genetic screening.36,37 The opinion of these consensus conferences was that there is insufficient evidence at present to recommend universal population-based genetic screening for HH. However, both statements specified the urgent need for research on various aspects of HH and there was recognition that future studies may strengthen the case for genetic screening. The decisions were influenced by the uncertainty surrounding penetrance of HFE mutations, the optimal care of asymptomatic people carrying HFE mutations for the spectrum of conditions implicated in HH, and concer ns regarding possible stigmatisation and discrimination. The Australian perspective The frequency of C282Y homozygosity in the Australian population is among the highest in the world due to the high proportion of people of northern European descent. Australians appear to have a higher average iron load, possibly due to dietar y and other lifestyle factors.20 These f actors underscore the need for local research into the natural history and potential modifiers of disease expression in individuals at genetic risk. There are several research projects under way in Australia that will provide data relevant to decision-making for screening for HH. The Busselton study has already provided data on penetrance of HH, but detected too few C282Y homozygotes to examine modifiers of penetrance. A large pilot screening study, Haemscreen, is under way through the Murdoch Institute in Melbourne and aims to screen 10,000 young adults between the ages of 18 and 35 in workplace settings. One of the study’s major aims is to assess the acceptance of genetic screening for haemochromatosis. To date the Haemscreen study has screened 1,000 people and detected two previously undiagnosed C282Y homozygotes who have signif icant iron overload (personal communication Dr Katrina Allen). Attitudes to genetic testing and sensitivities surrounding the possibility of discrimination about insurance may be different in Australia. As Australia has universal health care coverage and insurance companies must provide health insurance at one rate to all (community loading), concerns regarding health insurance are VOL. and specif icity of the phenotypic tests, estimates of disease penetrance and potential adverse effects. A recent Australian study found that testing for TS followed by a conf irmatory test and genetic testing of persistently elevated cases were the cheapest approaches.35 Testing the family members of affected cases also increased the cost effectiveness of screening. The study concluded that costs for screening using TS could be recovered if complications of HH were prevented by early detection and treatment. Without precise information on age-specif ic penetrance, it is difficult to determine whether the benef its of screening outweigh the costs and risks because low penetrance means low specificity and a high false-positive rate. Are there adequate facilities for diagnosis and treatment of HH? Implementation of a screening program for HH would require substantial resources. In addition to set-up costs, all par ticipants AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 26 NO. 6 Policy Public health aspects of genetic screening Recommendations 1. To take a national approach to addressing genetic screening for hereditary haemochromatosis (HH). 2. To identify experts, individuals, and groups within Australia who are active in the following areas: • clinical, epidemiolo gical or public health related research into HH; • relevant consumer groups; and • genetic testing for genes involved with HH. 3. To facilitate communication between the persons and groups identif ied above and to develop local expertise and knowledge with respect to genetic screening for HH and scientif ic data relevant to Australia. 4. To establish a means of monitoring developments arising from local and inter national research that may impact on the issue of genetic screening for HH. 5. To establish a laborator y-based sentinel surveillance system to monitor genetic testing for HFE mutations. Such a system could be modelled on sentinel clinics for infectious diseases and could also be extended in the future to monitor genetic testing for other conditions as tests become available. 6. To support and encourage local research addressing issues relevant to screening for HH, such as: • Penetrance, natural histor y and modifiers of disease expression in individuals at risk of HH. • Effectiveness of genetic screening in reducing morbidity and mortality. • Cost effectiveness of genetic screening. • Psycho-social aspects of the genetic screening process, in particular assessing potential adverse effects for persons identified with mutations. • Feasibility and possible protocols for genetic screening. 7. To promote and support education of health professionals and the general public regarding screening for HH. 8. To begin assessment of the potential impact on health services of population-based screening for HH. diminished, 34 although a waiting period may be required for new private health policies. As mentioned earlier, concer ns regarding life insurance are being addressed in Australia. 34 The cost effectiveness of a screening program may be greater in Australia because disease expression in those at genetic risk may be higher than in other countries, the higher prevalence of C282Y homozygosity means a greater proportion of people are genetically susceptible to HH, and costs of screening and treatment may be lower than in other countries, especially the US. Recommendations HH meets many of the public health criteria for screening, as summarised in Table 1. However, there is an ur gent need for local data relating to penetrance of the common HFE genotypes, natural history of early stage HH, effectiveness of screening in reducing morbidity and mortality from disease, potential adverse effects of screening and optimal screening strategies. In particular, there is a lack of knowledge about the age-specific risks of various conditions by genotype, although recent data suggest that clinical penetrance of disease may be ver y low.18 Data on penetrance and its modifiers will be forthcoming from the large ongoing NIH study. Additional data from well-designed epidemiological studies are needed to quantify the risk associated with different environmental, genetic and dietary factors in genetically susceptible individuals. An area of universal agreement in the literature is the need for education of health professionals, particularly GPs, about HH specifically and genetic testing in general. A heightened awareness of HH will increase in targeted testing in general practice for patients presenting with symptoms related to iron overload such as lethargy, arthralgia and type 2 diabetes, although the predictive value and cost-effectiveness of this approach is unclear. It is 2002 VOL. 26 NO . 6 important to monitor the uptake of genetic testing in the medical community for future planning and assessing the adequacy of genetic counselling. This could be done using a laboratory-based sentinel surveillance system, similar to that established for HIV testing, that could provide data regarding reasons for testing on a regular basis. Education of the public will also be important but should be approached cautiously, and only once there is consensus in the medical and public health communities. The momentum may gather for population-based screening depending on the results of research studies that are addressing outstanding research questions regarding penetrance and natural history. It is important that the public health community plays an active role in decision-making regarding screening. Facilitating dialogue between researchers, clinicians and public health professionals is an important way of advancing the debate around genetic screening for HH and other conditions in Australia. Acknowledgements This work was supported by the Victorian Health Promotion Foundation, the Department of Human Ser vices Victoria, and the National Health and Medical Research Council. We thank Tessa Keegel for manuscript editing. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Australian and New Zealand Journal of Public Health Wiley

Public health aspects of genetic screening for hereditary haemochromatosis in Australia

Loading next page...
 
/lp/wiley/public-health-aspects-of-genetic-screening-for-hereditary-gaxOHV0sHK

References (88)

Publisher
Wiley
Copyright
Copyright © 2002 Wiley Subscription Services
ISSN
1326-0200
eISSN
1753-6405
DOI
10.1111/j.1467-842X.2002.tb00360.x
Publisher site
See Article on Publisher Site

Abstract

Abstract Hereditary haemochromatosis (HH) is an inherited disorder of iron absorption. It meets several of the key public health principles for population-based screening and is considered to be a test-case for public health genetics. However, there has been relatively little debate in the public health or wider community regarding the merits of population-based genetic screening for HH. Genetic susceptibility to HH occurs in about 1:200 people and although mortality is low (age-standardised rate 2.75/million), there are potentially serious clinical manifestations of iron overload. Regular venesection is a simple and effective treatment for early stage iron overload. DNA-based testing is available and iron overload may be identified using serum transferrin saturation and ferritin tests. However, there are impor tant gaps in knowledge relevant to screening for HH. The limited data on penetrance of HFE genotypes, and thus the uncer tain probability that genetically susceptible individuals will develop clinically significant disease, is a major impediment to population-based genetic screening. Clinical evidence supports treating ear lystage disease but no randomised controlled trials of the effectiveness of screening in reducing the burden of disease have been conducted. In addition, the natural history of early stages of HH and factors that may modify progression are unclear. Two international consensus panels on HH concluded that there is insufficient evidence for population-based screening at present. We present recommendations to advance the debate on screening for HH in Australia. (Aust N Z J Public Health 2002; 26: 518-24) Dorota M. Gertig, Ashley Fletcher and John L. Hopper Centre for Genetic Epidemiology School of Population Health, , University of Melbour ne, Victoria ereditary haemochromatosis (HH) is a disorder of iron overload that may occur in individuals who have an inherited susceptibility due to mutations in the HFE gene. Increased absor ption of dietary iron leads to a potentially har mful accumulation of iron in parenchymal cells of various organs and tissues, in particular the liver, heart and pancreas. Two common mutations (C282Y and H63D) account for more than 90% of all cases of HH. The onset of disease is age-dependent and usually manifests in middle age with non-specif ic symptoms such as f atigue, arthralgia, impotence and arthritis. As iron overload increases, symptoms may become more severe and in later life it may lead to severe disease that includes liver cirrhosis, cardiomyopathy and diabetes.1 The classic picture of HH, known as ‘bronzed diabetes’ is a rare condition that results from severe or gan damage due to iron deposition. Since the identification of the HFE gene mutations responsible for HH in 1996,2 the prospect of population-based genetic screening for this condition has been debated. HH fulf ils many of the public health criteria for diseases suitable for screening. In fact, it has been proposed that screening for HH may become a model for future population-based genetic screening of other hereditar y conditions as appropriate DNA-based tests and therapeutic inter ventions become availab le.3 Therefore, it is par ticularly important to ensure that the public health implications of a screening program for HH are thoroughly considered. There has been substantial debate in the literature regarding the optimal screening approach for HH.4 Genetic screening refers to testing individuals directly for genetic susceptibility using a DNA-based test. Although this approach will detect a proportion of people at an earlier stage of disease prior to development of iron overload, it is unclear how many genetically susceptible people will go on to develop clinically signif icant disease. Phenotypic screening refers to identifying individuals with iron overload initially using transfer rin saturation (TS) and serum fer ritin (SF) tests. The sensitivity and specificity depend on the cut-off used and phenotypic screening will detect people who have iron overload and def iciency from other causes.5 However, the natural histor y of iron overload is uncertain and issues such as laboratory standardisation, education of medical practitioners and cost-effectiveness still need to be resolved. Another approach is case-f inding through testing the relatives of people with HH or people with symptoms or conditions that may be related to iron overload, such as f atigue, diabetes, etc. While screening of relatives is widely recommended, there are at present few data on the predictive value of various symptoms related to iron overload in detecting HH. Because HH is considered to be a test-case for public health genetics, in this review we concentrate on public health aspects of genetic screening for HH. Public health criteria Screening programs for all diseases, both hereditary and non-hereditary, should meet Submitted: May 2002 Revision requested: July 2002 Accepted: September 2002 Correspondence to: Dorota Gertig, Centre for Genetic Epidemiology, School of Population Health, University of Melbourne, 723 Swanston Street, Carlton, Victor ia 3053. Fax: (03) 9347 6929; e-mail: d.gertig@unimelb.edu.au AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH VOL. 26 NO. 6 Policy Public health aspects of genetic screening Table 1: Public health criteria for screening. Public health criteria The condition The condition should be an important health problem. The natural histor y of the disease should be understood. Hereditary haemochromatosis should be recognisable at an early stage. The treatment Treatment at an early stage should be more effective than treatment at a late stage. The test There should be a suitable test, in ter ms of sensitivity, specificity and positive predictive value, and the test should be acceptable to the population. For diseases of insidious onset, screening should be repeated at intervals deter mined by natural histor y. The screening program Physical and psychological harm to those screened should be less than the benefit. There should be adequate facilities for diagnosis and treatment. Does genetic screening for HH meet these criteria? Yes. History of early stage disease not well understood. Penetrance uncer tain. Iron overload detectable ear ly but case definition complicated by uncertain natural histor y and non-specific clinical picture. Yes, based on Level III and IV evidence. Yes. Laboratory standardisation of transferrin saturation (TS) and serum ferritin (SF) an issue. No need to repeat genetic screening test but TS and SF must be monitored in C282Y homozygotes. Insufficient evidence. No RCTs of screening have been conducted. Yes, potentially, but depends on the extent of screening and treatment. defined criteria to assess suitability and appropriateness, in particular that: • screening must result in more benef it to the targeted population than harm; and • early treatment, initiated because of screening, is shown to be effective in reducing morbidity or mortality from the disease.6 The public health issues for genetic screening (see Table 1) are similar to screening for non-hereditary diseases, although the terminology differs and there are some special considerations particularly with respect to implications for f amily members. First, genetic screening does not identify individuals who actually have disease; rather, the presence of a genetic prof ile associated with a particular disease only indicates the individual is at increased risk. This is similar to screening for elevated cholesterol as a risk factor for heart disease. The probability that an individual with a particular genetic profile will develop an outcome of interest (e.g. a specif ic disease or set of symptoms) by a cer tain age (cumulative age-specif ic risk) is referred to as the ‘penetrance’ of the genotype for that outcome. For most hereditary disorders the penetrance is ‘incomplete’, that is the chance of developing a disease is not 100% and it may be modif ied by environmental and other genetic factors. For example, different mutations in the BRCA1 and BRCA2 genes are associated with different lifetime risks of breast, ovarian and other cancers. The penetrance of these mutations appears to vary by the position of the mutations within the gene and by other genetic and environmental influences.7 Second, the potential adverse effects of ‘labelling’ a person with a genetic susceptibility when they have not yet developed or may never develop disease, must be addressed. These include psychological effects and the potential for discrimination. Third, genetic screening has consequences for family members. If the decision is made to inform family members of individuals found to have genetic risk, the provision of appropriate counselling and other issues must be considered. 2002 VOL. 26 NO . 6 Is HH an important health problem? Table 2 shows that the C282Y and H63D mutations are rare in Aboriginal and Asian populations and that the C282Y frequency is higher in northern than southern European populations. The frequency of these mutations in Australia and New Zealand is most comparable to nor thern European populations so that the C282Y allele frequency translates to between 1:150 and 1:300 being homozygotes. The estimated proportion of HH due to C282Y homozygosity (two copies of the C282Y mutation, one from each parent) ranges from 33% in southern Italy, 69% in northern Italy, 8 and up to 100% in Australia and the United Kingdom.9,10 Because homozygosity for the H63D mutation appears to account for a much smaller proportion of HH than the C282Y mutation,11 screening for that mutation has not generally been advocated. Haemochromatosis due to causes other than HFE mutations is rare; therefore the relative risk of HH due to HFE mutations is high. The odds of dev eloping iron ov erload for C282Y homozygotes compared with the odds of developing iron overload in controls are very high (estimates of odds ratio (OR) ~2,000 and 4,000 respectively). 11 Reported risks for other genotypes are much smaller: OR=25 and 32 for genotype C282Y/H63D; OR=8.6 and 5.7 for H63D/H63D. 11,12 The validity of these estimates is uncertain due to the v aried inclusion criteria for cases, the range of case def initions and a lack of comparable control groups in several of the studies. In 1998, the age-standardised rate of haemochromatosis-related deaths in Australia was 2.75 per million, compared with 1.8 per million in the US in 1992.13 Mor tality is likely to be underestimated due to the variety of clinical manifestations from the disease and under-diagnosis of the condition. Is the natural history of HH understood? The natural history of HH has been studied predominantly AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Gertig, Fletcher and Hopper Article Table 2: HFE genotype and allele frequencies from Australian and New Zealand studies, with summaries of other regions. Year Australia 199738 1998 Size Study design Source C282Y C282Y 0.5% 0% 0.5% 0% C282Y – 11.2% 1.1% 11.9% 13.0% H63D H63D H63D – C282Y H63D C282Y (%) H63D (%) 1,660 Cross-sectional 185 3,011 69 Cross-sectional Prospective Casecontrol Newbor ns Australian Aborigines General population Liver disease (not HH) General population 0.5% 2.1% 4.3% 3.8% 23.6% 20.3% 2.2% 1.4% 199914,60 199940 New Zealand 199741 199842 1,032 Cross-sectional 1.0% 0.5% 0.7% 0.02% 18.6% 11.4% 12.6% 3.4% 2.3% 2.2% 2.6% 22.6% 23.2% 23.3% 1.8% 2.4% 0.7% 10.3 7.0 (Range) 8.2 (6.2-9.9) 1.6 (0.2-3.7) 6.2 (5.3-6.6) 0.1 (0.0-0.2) 14.4 (Range) 15.0 (11.5-15.3) 12.6 (11.9-20.6) 15.1 (14.6-15.4) 0.4 1,064 Cross-sectional Electoral roll Europe (Summary) Northern 43-48 12,733 Southern 8,49-53 4,598 US (Summary) Caucasians17,54-56 12,534 Asia (Summary) Asians57-59 0.5% 9.3% 2.4% 23.2% 2.1% 1,350 0% 2.0% 0% 2.0% 0% among people with advanced disease. Cases identif ied because a relative has HH are more likely to be symptomatic due to the sharing of other genetic or environmental modifying factors.14 There are a few retrospective studies of people with HH, only some of which included individuals who were untreated,15 but these are not generalisable to population screening as the majority of people with HH had cir rhosis at presentation. The only pub lished prospective data of the natural history of iron overload come from the Busselton study in Western Australia 14 that included 12 untreated C282Y homozygotes (mean age 49, range 30 to 74), all of whom had elevated TS and more than half (58%) had elevated SF at the end of four years of follow up. HFE penetrance Estimates of penetrance of HFE genotypes depend on age, sex and the outcomes used to define penetrance, i.e. iron overload or various clinical disease endpoints. There are relatively few reliable estimates of penetrance of the C282Y homozygous genotype and, homozygotes had clinical features consistent with iron overload; however, the frequency of symptoms was not compared with a control series. In a primary-care population, Beutler and colleagues also reported a higher average iron index in homozygotes (n=152)17 but found the prevalence of clinical symptoms among C282Y homozygotes to be less than 1%.18 Thus the penetrance of C282Y homozygosity for iron overload may be approximately 50%, but the penetrance for clinical disease may be ver y low. Modifiers of HFE penetrance Genetic and environmental factors (including dietar y factors) that are known or likely to affect iron absorption or excretion may be potential modif iers of disease expression in genetically susceptible individuals.3 Both age and gender are important in general, these estimates are not age-specif ic. In a pooled analysis of case-control studies the estimated penetrance for iron overload in C282Y homozygotes was between 31 and 94%, depending on the prevalence of iron overload due to disease and the prevalence of C282Y homozygosity.11 An earlier pooled analysis of seven population-based studies16 estimated penetrance to be 50% for males and 44% for females. The Busselton study in Wester n Australia found penetrance of C282Y homozygosity for raised serum fer ritin was 58%14 based on 12 homozygotes. Four of these determinants of body iron stores. In the general population, iron levels increase with age and are lower in women due to loss of iron through regular menstruation and the effects of pregnancy.19 Dietary f actors, such as red meat intake and alcohol intake, have been shown to be predictors of serum fer ritin, but to different degrees for men and women.20,21 A recent study of Australian twin pairs found that after adjustment for age, gender and body mass index, HFE genotype explained only a small proportion of the variation in iron stores and about half the remaining variance in serum ferritin was accounted for by genetic f actors other than mutations in HFE.22 Animal models suggest that genes in the iron metabolism pathway may play a role in modifying HFE phenotype.23 Additional data will be forthcoming from a large study funded by the National Institutes of Health (NIH) in the US that VOL. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 26 NO. 6 Policy Public health aspects of genetic screening Figure 1: Age-specific haemochromatosis-associated mortality in Australia, by sex, 1997/98. 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85+ Ma les Fe ma les Data were obtained on the cause of death as presented on death cer tificates from the Austr alian Institute of Health and Welfare to estimate the mortality from HH in Austr alia. As there is no specific ICD9 code for haemochromatosis, the search was performed in accordance with previously established methods;13 briefly, iron disorders with known codes were subtracted from the general code 275.0 (disorders of iron metabolism, which includes haemochromatosis). There w ere 141 haemochromatosis-associated deaths (114 males and 27 females) in 1997 and 1998. Mortality Rate per Million Age is enrolling more than 100,000 people in primary care settings. This study will evaluate penetrance of HFE mutations, modifiers of penetrance and acceptability of screening. Is there a suitable test for HH and is the test acceptable to the population? A suitable screening test must be accurate and reliable. There are two screening tests for HH: a DNA-based test that identif ies people with genetic susceptibility and a phenotypic test that def ines iron overload. In 1998, testing for the C282Y mutation of the HFE gene became the f irst genetic test eligible for a Medicare Australia rebate.29 The test is reported to be highly accurate in detecting the C282Y and H63D mutations b ut sensitivity for predicting HH depends on the propor tion of HH due to the common HFE mutations in the screened population. For example, for southern Europeans this propor tion may be as low as 60%. Phenotypic testing involves assessment of TS and SF levels, which are relativel y inexpensive and ha ve reasonab le sensitivity and specificity, although elevated levels need to be followed up with a repeat fasting sample. Is HH recognisable at an early stage? The case definition for HH includes the presence of a deleterious HFE genotype and iron overload (as defined by raised TS and SF). Clinical signs and symptoms of conditions considered to be caused by iron overload may be present but are not necessarily part of the def inition and may be non-specif ic. Although early stages of iron overload are readily detectable using TS and SF, these levels are dependent on age and fasting status. In the Australian context approximately 10% of people with HH will not carry HFE mutations.24 Is treatment at an early stage more effective than treatment at a late stage? The standard treatment of early stage HH (i.e. before serious liver damage) is periodic phlebotomy. Treatment is usually lifelong, with on average four venesections per year needed to maintain normal iron levels. Annual assessment of TS and SF levels is recommended.25 At present, support for the effectiveness of early treatment is based largely on Level III and IV evidence (limited cohort and time series studies, and expert opinion, respectively). There have been no randomised controlled trials of the effectiveness of treatment for early stage HH. For patients with advanced HH, survival at five and 10 years is better than for untreated patients, after adjustment for age and clinical conditions.15,26 Studies have shown that survival in patients with HH who have undergone regular treatment, and who did not have cirrhosis or diabetes at presentation, is comparable to that of the general population.27,28 Therefore, the weight of clinical evidence suggests that earlier diagnosis and treatment of advanced HH improves survi val by several years and that individuals with HH who are treated at an early stage, i.e. do not have severe organ damage, will most likely have a normal life expectancy. Screening should be repeated at intervals determined by natural history As genetic testing is highly accurate and genetic susceptibility remains constant over a person’s lifetime, it should only be necessar y to monitor serum iron markers of individuals who are C282Y homozygotes (and possibly compound heterozygotes) at regular inter vals. The frequency of such monitoring will vary according to the rate at which de-ironing occurs. Physical and psychological harm should be less than the benefit The most important criterion for a screening program is that it results in sufficient benef it to offset any physical or psychological har m to the screened population, including their families. There should be evidence from high-quality randomised controlled trials (RCTs) that the screening program is effective in reducing mor tality and morbidity. In contrast to screening for other chronic conditions (e.g. mammographic screening for breast cancer), there have been no RCTs of screening for HH. Conduct of an RCT for HH would require a better understanding of the natural history of 2002 VOL. 26 NO . 6 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Gertig, Fletcher and Hopper Article iron overload, as it is unclear what the appropriate endpoints for such a trial would be. Therefore, at present there is inadequate evidence to assess whether screening results in more benefit to the population than harm. Some of the possible adverse effects of screening are false positives, false negatives, labelling, costs and premature implementation.30 Of par ticular concern, especially with respect to genetic screening, is that the screening program must be socially and ethically acceptable to health professionals and the public. Although false positives for genetic results are rare due to the high accuracy of DNA-based testing for specif ic mutations, the same does not apply to phenotypic screening. False positives may occur after follow-up testing for serum iron mark ers, although consistent false positive results are unlikely over time. False negative results from genetic testing may also occur as described above. Few studies have assessed the issues of labelling, stigmatisation and discrimination with respect to genetic screening for HH. In a small Canadian study it was found that genetic testing for HFE mutations was well accepted.31 On a f amily level there is the question of how other members react to a relative found to be at genetic risk. Issues range from the obligation of whether to tell other members, to others not wanting to be tested due to problems with health insurance. Data on the overall attitude of health insurers to people who are at genetic risk are not available, but there are individual case reports indicating that this may be a problem.32,33 An agreement has been made with the umbrella organisation for life and disability insurance in Australia (Investment and Financial Ser vices Association) and the Haemscreen study regarding life insurance policies for people undergoing screening for HH. This unique agreement will mean that homozygotes without organ damage and w ho have normal iron levels or whose iron levels retur n to normal following treatment will obtain policies at baseline rates.34 Cost effectiveness A number of factors affect the cost-effectiveness of screening, including the cost of genetic testing (which is likely to decrease over time as testing becomes more widespread), the sensitivity would require pre- and post-test genetic counselling as well as regular monitoring of af fected and unaffected C282Y homozygotes and treatment of iron overload in affected individuals. Based on an estimated prevalence of the C282Y genotype in people of northern European origin of 0.5%, 500 homozygotes would be detected for ever y 100,000 people screened. Using current estimates of penetrance, approximately half of the detected homozygotes over their lifetime would require treatment and regular monitoring; the other half would require monitoring of serum iron studies on an ongoing basis. The international perspective Since the characterisation of the HFE gene in 1996, there have been two major international consensus conferences examining the suitability of HH for population-based genetic screening.36,37 The opinion of these consensus conferences was that there is insufficient evidence at present to recommend universal population-based genetic screening for HH. However, both statements specified the urgent need for research on various aspects of HH and there was recognition that future studies may strengthen the case for genetic screening. The decisions were influenced by the uncertainty surrounding penetrance of HFE mutations, the optimal care of asymptomatic people carrying HFE mutations for the spectrum of conditions implicated in HH, and concer ns regarding possible stigmatisation and discrimination. The Australian perspective The frequency of C282Y homozygosity in the Australian population is among the highest in the world due to the high proportion of people of northern European descent. Australians appear to have a higher average iron load, possibly due to dietar y and other lifestyle factors.20 These f actors underscore the need for local research into the natural history and potential modifiers of disease expression in individuals at genetic risk. There are several research projects under way in Australia that will provide data relevant to decision-making for screening for HH. The Busselton study has already provided data on penetrance of HH, but detected too few C282Y homozygotes to examine modifiers of penetrance. A large pilot screening study, Haemscreen, is under way through the Murdoch Institute in Melbourne and aims to screen 10,000 young adults between the ages of 18 and 35 in workplace settings. One of the study’s major aims is to assess the acceptance of genetic screening for haemochromatosis. To date the Haemscreen study has screened 1,000 people and detected two previously undiagnosed C282Y homozygotes who have signif icant iron overload (personal communication Dr Katrina Allen). Attitudes to genetic testing and sensitivities surrounding the possibility of discrimination about insurance may be different in Australia. As Australia has universal health care coverage and insurance companies must provide health insurance at one rate to all (community loading), concerns regarding health insurance are VOL. and specif icity of the phenotypic tests, estimates of disease penetrance and potential adverse effects. A recent Australian study found that testing for TS followed by a conf irmatory test and genetic testing of persistently elevated cases were the cheapest approaches.35 Testing the family members of affected cases also increased the cost effectiveness of screening. The study concluded that costs for screening using TS could be recovered if complications of HH were prevented by early detection and treatment. Without precise information on age-specif ic penetrance, it is difficult to determine whether the benef its of screening outweigh the costs and risks because low penetrance means low specificity and a high false-positive rate. Are there adequate facilities for diagnosis and treatment of HH? Implementation of a screening program for HH would require substantial resources. In addition to set-up costs, all par ticipants AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 26 NO. 6 Policy Public health aspects of genetic screening Recommendations 1. To take a national approach to addressing genetic screening for hereditary haemochromatosis (HH). 2. To identify experts, individuals, and groups within Australia who are active in the following areas: • clinical, epidemiolo gical or public health related research into HH; • relevant consumer groups; and • genetic testing for genes involved with HH. 3. To facilitate communication between the persons and groups identif ied above and to develop local expertise and knowledge with respect to genetic screening for HH and scientif ic data relevant to Australia. 4. To establish a means of monitoring developments arising from local and inter national research that may impact on the issue of genetic screening for HH. 5. To establish a laborator y-based sentinel surveillance system to monitor genetic testing for HFE mutations. Such a system could be modelled on sentinel clinics for infectious diseases and could also be extended in the future to monitor genetic testing for other conditions as tests become available. 6. To support and encourage local research addressing issues relevant to screening for HH, such as: • Penetrance, natural histor y and modifiers of disease expression in individuals at risk of HH. • Effectiveness of genetic screening in reducing morbidity and mortality. • Cost effectiveness of genetic screening. • Psycho-social aspects of the genetic screening process, in particular assessing potential adverse effects for persons identified with mutations. • Feasibility and possible protocols for genetic screening. 7. To promote and support education of health professionals and the general public regarding screening for HH. 8. To begin assessment of the potential impact on health services of population-based screening for HH. diminished, 34 although a waiting period may be required for new private health policies. As mentioned earlier, concer ns regarding life insurance are being addressed in Australia. 34 The cost effectiveness of a screening program may be greater in Australia because disease expression in those at genetic risk may be higher than in other countries, the higher prevalence of C282Y homozygosity means a greater proportion of people are genetically susceptible to HH, and costs of screening and treatment may be lower than in other countries, especially the US. Recommendations HH meets many of the public health criteria for screening, as summarised in Table 1. However, there is an ur gent need for local data relating to penetrance of the common HFE genotypes, natural history of early stage HH, effectiveness of screening in reducing morbidity and mortality from disease, potential adverse effects of screening and optimal screening strategies. In particular, there is a lack of knowledge about the age-specific risks of various conditions by genotype, although recent data suggest that clinical penetrance of disease may be ver y low.18 Data on penetrance and its modifiers will be forthcoming from the large ongoing NIH study. Additional data from well-designed epidemiological studies are needed to quantify the risk associated with different environmental, genetic and dietary factors in genetically susceptible individuals. An area of universal agreement in the literature is the need for education of health professionals, particularly GPs, about HH specifically and genetic testing in general. A heightened awareness of HH will increase in targeted testing in general practice for patients presenting with symptoms related to iron overload such as lethargy, arthralgia and type 2 diabetes, although the predictive value and cost-effectiveness of this approach is unclear. It is 2002 VOL. 26 NO . 6 important to monitor the uptake of genetic testing in the medical community for future planning and assessing the adequacy of genetic counselling. This could be done using a laboratory-based sentinel surveillance system, similar to that established for HIV testing, that could provide data regarding reasons for testing on a regular basis. Education of the public will also be important but should be approached cautiously, and only once there is consensus in the medical and public health communities. The momentum may gather for population-based screening depending on the results of research studies that are addressing outstanding research questions regarding penetrance and natural history. It is important that the public health community plays an active role in decision-making regarding screening. Facilitating dialogue between researchers, clinicians and public health professionals is an important way of advancing the debate around genetic screening for HH and other conditions in Australia. Acknowledgements This work was supported by the Victorian Health Promotion Foundation, the Department of Human Ser vices Victoria, and the National Health and Medical Research Council. We thank Tessa Keegel for manuscript editing.

Journal

Australian and New Zealand Journal of Public HealthWiley

Published: Jan 1, 2002

There are no references for this article.