|
Screening for hemochromatosis in children of homozygotes: prevalence and cost-effectiveness |
Adams P C, Kertesz A E, Valberg L S |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology Screening for hemochromatosis with transferrin saturation as the initial screening test, which is repeated in a fasting state if abnormal findings result from the former. Children with abnormal findings in the latter test, then undergo a serum ferritin test undertaken on the same blood sample. Those patients with high transferrin saturation and abnormal findings from ferritin, undergo a diagnostic liver biopsy with measurement of the hepatic iron concentration. Finally, those with a high hepatic iron index begin therapeutic venesections. For those having an abnormal finding from transferrin saturation, but with a normal serum ferritin, tests are performed "at 2- to 4-year intervals for up to 20 years, and if the ferritin becomes abnormal, they proceed to liver biopsy or diagnostic venesections".
Type of intervention Screening, diagnosis and treatment.
Study population Children of homozygous parents.
Setting Hospital and primary care. The economic study was carried out in Ontario, Canada.
Dates to which data relate The data for the prevalence and part of the effectiveness analysis were collected from 1965 to 1995. These data were supplemented with data from the literature "as necessary", from studies published between 1981 and 1993. The resource use data were obtained partly from the same period and partly from a study published in 1995. The price year was 1994.
Source of effectiveness data Effectiveness data were derived from a single study, review of previously completed studies, and authors' assumptions.
Link between effectiveness and cost data The costing related to the intervention was retrospectively performed on the same patient sample as that used in the effectiveness analysis. The costing related to the no-screening option was based on the results of a previous decision analytic model carried out by the authors.
Study sample Power calculations were not used to determine the sample size. The study sample consisted of 255 individuals ranging in age from 7 to 73 years.
Study design Prospective case series, carried out in a single centre. The duration of the study was 30 years. No loss to follow-up was reported.
Analysis of effectiveness The principle (intention to treat or treatment completers only) used in the study was not explicitly specified. The prevalence of hemochromatosis and the cases of life-threatening complications were reported.
Effectiveness results The prevalence of hemochromatosis was 0.043 (11 cases out of 255). There was one case of a life-threatening complication (cirrhosis).
Clinical conclusions "The actual prevalence of hemochromatosis in this study was 11 of 255 = .043, which is similar to the predicted prevalence of .05".
Modelling A decision tree was used to estimate costs and benefits.
Outcomes assessed in the review The main outcomes assessed in the review were prevalence, symptom onset, symptom-specific survival, penetrance of disease, sensitivity and specificity of transferrin saturation, serum ferritin, and compliance with liver biopsy and follow-up venesections.
Study designs and other criteria for inclusion in the review Sources searched to identify primary studies Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data The methods used to judge relevance and validity were not reported. The data were extracted by means of summary statistics.
Number of primary studies included At least nine studies were included.
Methods of combining primary studies Investigation of differences between primary studies Results of the review The main values obtained from the review and used in the economic analysis were as follows:
predicted prevalence of hemochromatosis amongst children of an affected parent, 0.05;
normal life expectancy in males was 72 years and 78 for females;
average age of onset of life-threatening symptoms in males and females, 54 and 59 years;
death from heart failure in males and females, 55 and 60 years;
death from cirrhosis in males and females, 67 and 72 years;
death from diabetes in males and females, 68 and 73 years;
death from hepatocellular carcinoma in males and females, 64 and 69 years;
sensitivity and specificity of transferrin saturation (TS) test (for cut off points of greater than 50% for the initial test and greater than 50% for females and greater than 60% for males in the fasting sample), 0.92 and 0.93, respectively;
sensitivity and specificity of the serum ferritin 'alone' test with respect to transferrin saturation combined with serum ferritin (for cut off points as follows:Females aged 10 to 19, 40microg/L; aged 20 to 29, 65microg/L; aged 30 to 40, 80 microg/L. Males aged micro10 to 19, 100microg/L, aged 20 and more, greater than 350 microg/L.), 0.85 and 0.95, respectively;
compliance with liver biopsy and follow-up venesections, 90%.
Methods used to derive estimates of effectiveness Estimates of effectiveness were also derived from authors' assumptions.
Estimates of effectiveness and key assumptions Utilities assumed for the health states of the presence of cirrhosis, diabetes, heart failure, cirrhosis and diabetes, cirrhosis and heart failure, diabetes and heart failure, and cirrhosis and diabetes and heart failure were 0.8, 0.9, 0.5, 0.72, 0.78, 0.87, and 0.70, respectively. It was assumed that the presence of a hepatic iron index greater than 1.9 was a definitive evidence of hemochromatosis
Measure of benefits used in the economic analysis Quality-adjusted life days (QALDs) gained was the benefit measure used in the economic analysis. A decision tree was used to estimate the benefits, given the long term implications of the strategies. The measurement (global quality of life scores) was carried out in a previously published study.
Direct costs Costs were discounted. Quantities were not reported. The cost items were not fully reported separately. The costs measured were operating costs and costs of complications (cirrhosis, diabetes and heart failure), the cost analysis being performed from the perspective of a third-party payer. The estimation of resource use was based on actual data and authors' assumptions and using a decision model. The unit costs were based on the home institution's costs. The prices used were those prevailing in 1994. The costs of non-life threatening disorders (arthritis, impotence, and infection) were not included in the analysis.
Currency Canadian dollars (Can$). The exchange rate was reported to be Can$1 = US$0.72.
Sensitivity analysis The variables explored in the sensitivity analysis were the discount rate, prevalence rate, the age at screening, the sensitivity and specificity of screening tests (two sets of values obtained from the literature were used), and the cost of venesections. In addition, the results were presented for different values of age of screening by gender. One-way simple sensitivity and threshold analyses were carried out.
Estimated benefits used in the economic analysis The screening strategy yielded an expected gain, for a male aged 10 years, of 12 QALDs. A female aged 10 years was estimated to have an expected gain of 7 QALDs. The overall expected gain per patient was estimated to be 10 QALDs for a 10 year-old patient, and 14 QALDs for a 20 year-old.
Cost results The costs were discounted at 3%. The total net cost for the screening (relative to the no-screening strategy) was -Can$53 for a male of 10 years, in 1994 prices. The expected value of cost of screening for a 10 year-old female was estimated to be Can$30, at 1994 prices. The overall expected cost of the intervention was estimated to be Can$12for a 10 year-old child and $Can65 for a 20 year-old in 1994 prices.
Synthesis of costs and benefits The costs and benefits were only combined for the case of a 10 year-old female, since the remaining cases were shown to result in the intervention being the dominant strategy. The discounted cost, in 1994 prices, per QALD gained was used as the outcome measure for the synthesis. This was done using a 3% discount rate for costs. The intervention resulted in an expected cost in 1994 prices per QALY gained, for the above-mentioned case, of Can$1,479. The sensitivity analysis showed that the threshold value (above which the intervention turns out to be the dominant strategy for 10 year-old children) for the prevalence rate was 0.042. The threshold value (below which the intervention was the dominant strategy) for venesection cost was $30 and $40 for children of ages 10 and 20 years (baseline:$25). The threshold (below which dominance results) value for the discount rate was 0.047. By using values from the literature for transferrin saturation (sensitivity changing from 0.92 to 0.82, specificity from 0.93 to 0.88) and for transferrin saturation combined with serum ferritin (sensitivity changing from 0.94 to 0.85, specificity from 0.86 to 0.95), the cost per QALY gained ratio (for a 10 year-old child) at 3% discount rate expressed in 1994 prices was Can$3,051.
Authors' conclusions The authors concluded that:"Because the prevalence of hemochromatosis is higher in children of homozygotes than in the general population, screening with transferrin saturation and ferritin as early as age 10 years is recommended. Savings are augmented if the cost per venesection is eliminated by allowing hemochromatosis patients to become voluntary blood donors".
CRD COMMENTARY - Selection of comparators The reason for the choice of comparator (the 'no screening' option) is clear. Validity of estimate of measure of benefit The internal validity of the estimates of measure of benefits cannot be fully assessed because of the lack of information regarding the literature review and quality assessment of the primary studies included. Validity of estimate of costs Resource quantities were not fully reported separately from the costs and insufficient details were provided of the methods of cost estimation. The costs of non-life threatening complications and indirect costs were not included in the analysis. Other issues In view of the apparent lack of a systematic literature review, quality assessment of the primary studies included, and the statistical analysis of the costs, the results may need to be treated with some caution. The issue of generalisability was addressed in terms of the prices prevailing outside Canada and the race of the population considered, the prevalence value used being that found in "white populations of European descent", although the authors made clear that "these estimates do not apply to marriages of mixed race that are increasingly prevalent in modern urban centres". The authors acknowledged that the application of US prices may generate a different set of results. Source of funding Supported by Canadian Liver Foundation, Toronto, Ontario Canada, PSI Foundation, Toronto, Ontario, Canada, and James Morton Estate, London, Ontario, Canada.
Bibliographic details Adams P C, Kertesz A E, Valberg L S. Screening for hemochromatosis in children of homozygotes: prevalence and cost-effectiveness. Hepatology 1995; 22(6): 1720-1727 Indexing Status Subject indexing assigned by NLM MeSH Adolescent; Adult; Age Factors; Aged; Child; Cost-Benefit Analysis; Female; Ferritins /analysis; Hemochromatosis /epidemiology /genetics /prevention & Homozygote; Humans; Liver /pathology; Male; Mass Screening /economics; Middle Aged; Phlebotomy /economics; Quality of Life; Transferrin /analysis; control AccessionNumber 21996000138 Date bibliographic record published 31/07/1999 Date abstract record published 31/07/1999 |
|
|
|