|Costs and outcomes of early versus delayed migraine treatment with sumatriptan
|Halpern M T, Lipton R B, Cady R K, Kwong W J, Marlo K O, Batenhorst A S
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.
The use of early sumatriptan treatment for mild migraine versus delayed sumatriptan treatment for moderate to severe migraine. Doses of 50 mg or 100 mg were considered for both treatments.
Economic study type
The study population comprised a hypothetical cohort of patients experiencing mild migraine.
The setting was primary care. The economic study was conducted in the USA.
Dates to which data relate
The effectiveness and resource use data were derived from studies published between 1996 and 2000. The price year was 2000.
Source of effectiveness data
The effectiveness evidence was derived from a review of completed studies.
A decision tree model was constructed to assess the costs and benefits of the two alternative treatment strategies for migraine. Two separate trees were considered for the two strategies.
In the tree for early treatment with sumatriptan, after 2 hours, the patients could:
have resolved their headache and be pain free (but could experience recurrence and require further treatment);
progress to persistent mild pain; or
develop moderate/severe pain, thus requiring another dose of the same or an alternative medication (thus resolving or not).
For patients who did not resolve, emergency room (ER) or physician visits were considered, while a proportion of patients did not seek medial intervention. The status of the patients was re-evaluated 4 hours after the initial therapy.
The tree for delayed treatment was similar, but the patients did not treat initial mild pain and received sumatriptan only in the case of persistent mild pain or progression to moderate to severe pain (evaluated at 2 and 4 hours).
Outcomes assessed in the review
For early treatment, the outcomes estimated from the literature were the probabilities of the following with a dose of 50 or 100 mg:
pain free at 2 hours;
moderate/severe pain at 2 hours;
persistent mild pain at 2 hours;
for those with mild pain at 2 hours, re-treat with sumatriptan or with other medications, or do not treat;
for those with mild pain at 2 hours, outcomes of pain free at 4 hours, mild pain at 4 hours, moderate/severe pain at 4 hours;
for those with moderate/severe pain at 2 hours, re-treat with sumatriptan or with other medication, or do not treat;
for those with moderate/severe pain at 2 hours, outcomes of pain free at 4 hours, mild pain at 4 hours, and moderate/severe pain at 4 hours; and
For delayed treatment, the probabilities of the following were assessed:
to become pain free without treatment, progress to moderate/severe pain, have persistent mild pain;
for patients progressing to moderate/severe pain before treatment, outcomes of pain free at 2 hours, moderate/severe pain at 2 hours, mild pain at 2 hours;
for those with mild pain at 2 hours, re-treat with sumatriptan or with other medications, or do not re-treat;
for those with mild pain at 2 hours, outcomes of pain free at 4 hours, mild pain at 4 hours, and moderate/severe pain at 4 hours;
for those with moderate/severe pain at 2 hours, re-treat with sumatriptan or with other medications, or do not re-treat;
for those with moderate/severe pain at 2 hours, outcomes of pain free at 4 hours, mild pain at 4 hours, and moderate/severe pain at 4 hours; and
recurrence with or without treatment.
Study designs and other criteria for inclusion in the review
A systematic review does not appear to have been conducted. Clinical trials were used to provide the evidence.
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
Not stated. However, some of the inputs were obtained from values calculated from the primary data extracted from the trials.
Number of primary studies included
Three primary studies provided the effectiveness evidence.
Methods of combining primary studies
Average values were used, whenever possible, to combine the primary estimates.
Investigation of differences between primary studies
Results of the review
The probability values presented here are for doses of 50 mg (value for 100 mg dose in brackets). A single value refers to both dosages.
For early treatment, the probabilities were:
0.505 (0.670) for pain-free at 2 hours;
0.330 (0.240) for moderate/severe at 2 hours;
0.165 (0.090) for persistent mild pain at 2 hours;
for those with mild pain at 2 hours, 0.188 for re-treat with sumatriptan, 0.1 for re-treat with other medications, and 0.712 for do not treat;
for those with mild pain at 2 hours, 0.838 (0.855) for pain free at 4 hours, 0.054 (0.040) for mild pain at 4 hours, and 0.108 (0.105) for moderate/severe pain at 4 hours;
for those with moderate/severe pain at 2 hours, 0.415 for re-treat with sumatriptan, 0.294 for re-treat with other medication, and 0.291 for do not treat;
for those with moderate/severe pain at 2 hours, 0.474 (0.638) for pain-free at 4 hours, 0.175 (0.099) for mild pain at 4 hours, and 0.350 (0.263) for moderate/severe pain at 4 hours; and
0.13 for recurrence.
For delayed treatment, the probabilities were:
0.140 for become pain-free without treatment, 0.580 for progression to moderate/severe pain, 0.280 for have persistent mild pain;
for patients progressing to moderate/severe pain before treatment, 0.290 (0.360) for pain-free at 2 hours, 0.473 (0.465) for moderate/severe pain at 2 hours, 0.237 (0.175) for mild pain at 2 hours;
for those with mild pain at 2 hours, 0.188 for re-treat with sumatriptan, 0.1 for re-treat with other medications, and 0.712 for do not re-treat;
for those with mild pain at 2 hours, 0.455 (0.479) for pain free at 4 hours, 0.182 (0.142) for mild pain at 4 hours, and 0.363 (0.379) for moderate/severe pain at 4 hours;
for those with moderate/severe pain at 2 hours, 0.415 for re-treat with sumatriptan, 0.294 for re-treat with other medications, and 0.291 for do not re-treat;
for those with moderate/severe pain at 2 hours, 0.258 (0.357) for pain free at 4 hours, 0.247 (0.175) for mild pain at 4 hours, and 0.495 (0.467) for moderate/severe pain at 4 hours; and
0.180 for recurrence with treatment and 0.330 for recurrence without treatment.
Measure of benefits used in the economic analysis
A series of benefits measures were derived from the decision model. More specifically, the pain-free rates at 2 and 4 hours and the sustained pain-free rate, and the weighted disability values and cumulative disability hours (disability score multiplied by the amount of time spent with that score during the first 4 hours after treatment). The main benefit measure, which was then combined with the costs, was the sustained pain-free rate.
Discounting was not relevant since the costs were incurred during a short time. The unit costs were presented separately from the quantities of resources used. The health services included in the economic evaluation were drugs, physician visits and ER visits. The cost/resource boundary of society was adopted. The costs and rates of resource use were derived from published studies and wholesale prices (drugs). The costs associated with the long-term management of migraine and diagnostic imaging were not considered. The price year was 2000.
Statistical analysis of costs
The costs were treated deterministically.
The indirect costs were included since a societal perspective was adopted. The costs related to absence from work and decreased productivity in the workplace were estimated. The average earnings were estimated from the US Bureau of Labor Statistics, assuming an 8-hour workday. Resource use was derived from actual trial data. Discounting was not relevant because of the short timeframe of the study. The unit costs and the quantities of resources used were presented separately. The price year was 2000.
Sensitivity analyses were conducted to assess the robustness of the estimated cost-effectiveness ratios to variations in the probability of becoming pain-free while delaying migraine therapy (delayed treatment model) and the rate of migraine recurrence. The ranges of values used were derived from the literature. Univariate and threshold analyses were conducted.
Estimated benefits used in the economic analysis
All the results favoured the group of patients in the early treatment group. In general, sumatriptan 100 mg performed better than sumatriptan 50 mg.
When using 50 mg sumatriptan, the estimated pain-free rates at 2 hours were 50.5% for early treatment versus 45% for delayed treatment. The corresponding rates when using 100 mg sumatriptan were 67% (early treatment) and 53.6% (delayed treatment), respectively.
The pain-free rates at 4 hours and sustained pain-free rates were depicted graphically only. It would appears that the rates were higher for early treatment than for delayed treatment.
The weighted disability values, which were only depicted graphically, showed better results for early treatment than for delayed treatment. The results for 100 mg sumatriptan were better than those for 50 mg sumatriptan.
The cumulative disability hours for early versus delayed treatment were 3.23 versus 5.62 with 50 mg sumatriptan, and 2.36 versus 5.27 with 100 mg sumatriptan.
The medical costs were comparable between the groups, independent of the dosages. However, the workplace costs were substantially lower with early rather than delayed treatment and a similar result was reached in those who received 50 mg or 100 mg sumatriptan. Therefore, the total costs were much lower with early rather than delayed treatment, regardless of sumatriptan dosage. The actual figures were not reported, they were only illustrated graphically.
Synthesis of costs and benefits
An incremental cost-effectiveness ratio was calculated to combine the costs and benefits. However, early treatment was generally more effective and less expensive than delayed treatment. There was one exception: the cost per additional sustained pain-free patient receiving early treatment with sumatriptan 50 mg was $2.24 compared with patients receiving delayed treatment with sumatriptan 50 mg.
The sensitivity analyses indicated that the main conclusions reached in the base-case analysis were not changed. Substantial, unrealistic variations were needed if delayed treatment were to become more cost-effective than early treatment.
Compared with delayed sumatriptan treatment, the early treatment of mild migraine with sumatriptan led to greater sustained pain-free rates, decreased disability, and lower costs.
CRD COMMENTARY - Selection of comparators
The authors discussed the rationale for the choice of the comparators. They stated that delayed treatment represented a standard approach, as clinicians usually directed patients to wait until migraine pain had progressed from mild to moderate or severe before initiating any treatment. However, only sumatriptan was considered. No comparison with other medications, such as non-triptans, was conducted. You should decide whether this is a valid comparator in your won setting.
Validity of estimate of measure of effectiveness
The effectiveness evidence came from published studies. However, a systematic review of the literature does not appear to have been conducted and the primary studies were, presumably, identified selectively. The three studies used in the analysis were clinical trials, which ensures the internal validity of the sources used. However, the authors did not discuss the comparability of the study groups in terms of the patient samples, outcome assessment and follow-up. The method used to combine the primary estimates was reported for some outcomes. The issue of uncertainty in the primary estimates was partially addressed by varying two model inputs in the sensitivity analysis. Some data were derived from self-reported measures, and this may have introduced some bias into the estimates used in the model. However, the authors noted that any bias would have affected the two groups in equal measure.
Validity of estimate of measure of benefit
Several benefit measure were derived from the decision model, but only the sustained pain-free rate was then used in the synthesis of the costs and benefits. These measures were specific to the disease considered in the study and could not be compared with the benefits of other health care interventions.
Validity of estimate of costs
The authors did not state explicitly which perspective was adopted in the study. However, a societal perspective was used to capture all the relevant categories of costs, which were incurred by different patients. Therefore, the indirect costs (i.e. productivity losses) were included in the analysis. The unit costs, quantities of resources used, price year and source of data were reported. This enhances the possibility of carrying out reflation exercises and replicating the study in other settings. However, the cost estimates were specific to the study setting and no sensitivity analyses were conducted. Further, no statistical tests on the costs or quantities were conducted. Some categories of costs were not considered because the authors focused the economic analysis on the evaluation of costs associated with a single migraine episode. Therefore, the estimated savings were conservative. Finally, in the delayed strategy, the patients did not take any medication before 2 hours. In real life, patients could experience acute migraine and be treated earlier. This scenario was not considered.
The authors reported the results of other studies that evaluated non-triptan therapies as well as sumatriptan treatment. However, they stated that this was the first study to evaluate delayed versus early treatment using sumatriptan. The issue of the generalisability of the study results to other settings was not addressed and sensitivity analyses were not conducted to address uncertainty in the results of the study. This affected the external validity of the analysis. The authors noted that the resource use data came from a source different from that used to derive the effectiveness data. Ideally, the two types of data should have come from the same study.
Implications of the study
These results should encourage physicians to treat migraines when pain is mild, rather than waiting for more severe stages. Future studies should evaluate the economic impact of early sumatriptan versus non-triptan therapy, as well as the impact of early treatment on the patients' functional status, quality of life and satisfaction.
Source of funding
Funded by a study contract from GlaxoSmithKline.
Halpern M T, Lipton R B, Cady R K, Kwong W J, Marlo K O, Batenhorst A S. Costs and outcomes of early versus delayed migraine treatment with sumatriptan. Headache 2002; 42(10): 984-999
Other publications of related interest
Cady RK, Sheftell F, Lipton RB, et al. Effect of early intervention with sumatriptan on migraine pain: retrospective analyses of data from three clinical trials. Clinical Therapeutics 2000;22:1035-48.
Cady RK, Lipton RB, Hall C, et al. Treatment of mild headache in disabled migraine sufferers: results of the Spectrum Study. Headache 2000;40:792-7.
Adelman JU, Sharfman M, Johnson R, et al. Impact of oral sumatriptan on workplace productivity, health-related quality of life, healthcare use, and patient satisfaction with medication in nurses migraine. American Journal of Managed Care 1996;2:1407-16.
Subject indexing assigned by NLM
Cost-Benefit Analysis; Decision Trees; Humans; Migraine Disorders /classification /drug therapy /economics; Probability; Recurrence; Serotonin Receptor Agonists /economics /therapeutic use; Sumatriptan /economics /therapeutic use; Time Factors; United States
Date bibliographic record published
Date abstract record published