|
The use of decision-analytical modelling in economic evaluation of patch testing in allergic contact dermatitis |
Rajagopalan R, Anderson R T, Sarma S, Retchin C, Jones J |
|
|
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 The use of patch testing for patients suspected of having allergic contact dermatitis (ACD).
Economic study type Cost-effectiveness analysis.
Study population All patients with a suspicion of contact allergy who exhibited at least moderate disease activity were included in the study. 71% of the sample were women and 87% were Caucasian. The mean age of study participants was 45.5 years. For patch-tested and non-patch-tested patient groups, the percentage of time with bothersome signs/symptoms prior to enrolment was 64% and 65%, respectively. The pre-study dermatitis duration was 10 months (patch-tested) and 12 months (median) (non-patch-tested). Patients were stratified according to whether they were treated in "low-utility" clinics (those which patch tested less than 50% of the patients in the study) and "high-utility" clinics (those who patch tested 50% or more of the patients in the study). Patients were also stratified by their self-rated severity of the disease (mild, moderate and severe).
Setting The setting was secondary care. The economic evaluation was carried out in North Carolina, USA.
Dates to which data relate Dates for effectiveness and resource use data ranged from October 1995 to April 1997. The price year was not specified: a single price year appears not to have been used.
Source of effectiveness data The effectiveness data were derived from a single study.
Link between effectiveness and cost data The costing was undertaken retrospectively on the same patient sample as that in the effectiveness study.
Study sample The method of sample selection was not reported. Power calculations appear to have been conducted, as it was reported that 31 to 76 patients were recruited in each of the 10 sites, although 50 patients were required from each site. 567 patients were recruited in total. 429 patients were finally evaluated (after 2 were lost to follow-up): 184 (43%) in the intervention group (11 mild, 73 moderate, 100 severe) and 245 (57%) in the control group (38 mild, 107 moderate, 100 severe). Patients who were ruled out for contact allergy by means other than patch testing within the first six months were excluded from the study, but the percentage was not reported.
Study design A multi-centre (10) observational study was conducted over a period of one year. Patients were assigned to one of 2 groups according to patch-test status: patch-tested at screening or within 8 days of enrolment, and non-patch-tested. 2 patients were lost to follow-up.
Analysis of effectiveness The analysis of effectiveness was based on intention to treat. Patients who had completed a pharmacoeconomic questionnaire and a dermatology-specific quality of life (DSQL) questionnaire at the screening and at 6 months (if healed at 6 months) or at 6 and 12-months (otherwise) were evaluated. The primary outcomes included diagnosis (evaluated as the proportion of patients with confirmed diagnosis, proportion of patients with positively identified allergen, rate of patients advised to avoid allergenic substances and the rate of patients who benefited by a change in lifestyle), and disease management (assessed by results of patch testing, investigator's global evaluation and the disease management summary). Secondary outcomes were quality of life (QoL) measures estimating the impact of allergic contact dermatitis (ACD) on the patient's quality of life. This was assessed through a validated dermatology-specific QoL (DSQL) instrument.
Effectiveness results There was a statistically significant difference between the patch-tested and non-patch-tested groups in terms of the time to obtain a confirmed diagnosis (medians 8 and 175 days, respectively, p<0.001). The advice to avoid substances based on the identification of allergen was given to 76% of patch-tested patients and 43% of non-patch-tested patients. As a result of life style changes, 66% of patch-tested and 51% of non-patch-tested patients reported a 75% or more improvement in their disease symptoms after 6 months. The use of patch testing among patients with moderate to severe ACD results in a greater QoL benefits at 6 months than when patch-testing was not used, despite the fact that patients who were not patch-tested had less severe ACD. The greatest QoL benefit from patch testing relative to non-patch-testing occurred in patients with recurrent and/or chronic ACD. Less benefit was found for patients with mild or moderate disease. In the cohort with chronic ACD, greater QOL benefit from patch testing was derived when it was performed early (within 8 days) rather than later.
Clinical conclusions Patch testing was associated with a reduction in time to confirmed diagnosis, an improvement of patients' QoL and faster improvement of symptoms, compared to non-patch-testing.
Modelling A decision tree model was used to evaluate the expected costs and effectiveness of patch testing in different patient populations and dermatology clinics.
Measure of benefits used in the economic analysis The measure of benefit used in the economic analysis was the patient's expected increase in QoL measured with the DSQL instrument. A decision tree model was used to evaluate expected effects. The duration of benefits was one year.
Direct costs Costs were not discounted due to the short time frame of the study (one year). Quantities and costs were not reported separately. The direct costs included the costs of diagnosis and treatment, including the costs of office visits, and drugs without generic substitution and supplies, but excluded the costs of patch testing and carrying out the patch test. The quantity/cost boundary adopted was that of the health service. The method used to estimate quantities and costs was based on data collection from the sample of patients from the clinical study, and modelling using the decision tree. The source of cost data was not reported. The quantity of resources was measured during the observational study. The price date was not specified, and prices appear to have been those at the trial date (October 1995 to April 1997). Prices were not reported as having been reflated.
Statistical analysis of costs The costs derived from the decision tree model were not treated stochastically.
Indirect Costs Indirect costs were not analysed.
Sensitivity analysis Sensitivity analyses were performed using the costs of generic drugs.
Estimated benefits used in the economic analysis For "low-utility" clinics at one year:
Expected QoL differences for the patch-tested group were (disease severity rating in parentheses): -0.48 (mild), -2.21 (moderate), and -6.61 (severe).
Expected QoL differences for the non-patch-tested group were, -0.68 (mild), -1.93 (moderate), and -2.69 (severe).
For "high-utility" clinics at one year:
Expected QoL differences for the patch-tested group were, -0.56 (mild), -2.00 (moderate), and -3.31 (severe).
Expect QoL differences for the non-patch-tested group were, -0.10 (mild), -2.05 (moderate), and -1.77 (severe).
The duration of benefits was one year.
Cost results For patch-tested patients, the expected costs at "low-utility" clinics at one year were estimated as (disease severity in parentheses): $1,094 (mild), $1,839 (moderate), and $801 (severe).
For the non-patch-tested group these costs were: $973 (mild), $692 (moderate), and $984 (severe).
For patch-tested patients, the expected costs at "high-utility" clinics at one year were estimated at: $670 (mild), $999 (moderate)and $999 (severe).
For the non-patch-tested group these costs were: $1,071 (mild), $906 (moderate), and $1,689 (severe).
The duration of quantities/costs used in the economic analysis was one year.
Synthesis of costs and benefits For patch-tested patients, the cost per unit change in QoL for "low-utility" clinics was (disease severity rating in parentheses): $2,280 (mild), $832 (moderate), and $121 (severe).
For the non-patch-tested group these costs were: $1,431 (mild), $359 (moderate), and $366 (severe).
For the patch-tested patients, the cost per unit change in QoL for "high-utility" clinics was: $1,196 (mild), $500 (moderate), and $302 (severe).
For the non-patch-tested group these costs were: $10,708 (mild), $442 (moderate), and $960 (severe).
The substitution of generic drugs and the addition of indirect costs within the sensitivity analysis did not change the results.
Authors' conclusions Patch testing is cost-effective and reduces the cost of therapy in patients with severe ACD. Greater improvements in quality of life were achieved in patients with recurrent/chronic ACD who were patch tested than in patients who were not patch tested in the same group.
CRD COMMENTARY - Selection of comparators The selection of the comparator was justified from the objectives of the study.
Validity of estimate of measure of benefit The analysis was based on a cohort study, which was appropriate for the study question. However, the validity could have been influenced by the presence of bias inherent in this type of study design (observational study). In addition the competence of the dermatologists (regarding the time related decision as to whether to employ patch testing) could also have influenced the effectiveness. The effect of the unbalanced samples from the different study sites/dermatologists in terms of the representativeness of those samples, was not analysed. The estimation of benefits was directly obtained from the effectiveness analysis and this choice was justified.
Validity of estimate of costs Some problems with the analysis of costs were as follows: no perspective for the cost analysis was specified and it was not clear who bore the different costs of diagnosis and treatment. The costs of the patch-test packages (varying from $97 to $520) were not included in the analysis. The labour costs of handling the different patch tests were not included in the analysis. These costs could influence the cost-effectiveness results. Costs and quantities were not reported separately, and only total costs were reported. The source of the costs was not reported. A statistical analysis of costs was not performed. The price year was not reported.
Other issues Comparisons with findings from other studies were discussed. The generalisability of the study results was not addressed.
Implications of the study The study demonstrates the use of decision modelling in an observational study to assess the cost-effectiveness of patch testing in ACD.
Source of funding Funded by Glaxo Wellcome Inc, USA.
Bibliographic details Rajagopalan R, Anderson R T, Sarma S, Retchin C, Jones J. The use of decision-analytical modelling in economic evaluation of patch testing in allergic contact dermatitis. PharmacoEconomics 1998; 14(1): 79-95 Indexing Status Subject indexing assigned by NLM MeSH Anti-Allergic Agents /economics /therapeutic use; Cost-Benefit Analysis; Decision Support Techniques; Dermatitis, Allergic Contact /diagnosis /drug therapy /economics; Female; Humans; Male; Middle Aged; Patch Tests /economics; United States AccessionNumber 21998008266 Date bibliographic record published 30/06/2001 Date abstract record published 30/06/2001 |
|
|
|