| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Evidence-based Practice |
1 From the Institute for Technology Assessment, Massachusetts General Hosp, 101 Merrimac St, 10th Floor, Boston, MA 02114 (M.T.B., J.L.B., G.S.G.); Dept of Radiology (M.T.B., J.L.B., G.S.G.) and Vincent Gynecology Service (K.B.I.), Massachusetts General Hosp, Harvard Med School, Boston; Dept of Health Policy and Management, Harvard School of Public Health, Boston, Mass (G.S.G.); Dept of Epidemiology and Biostatistics, Erasmus Univ Med Center, Rotterdam, the Netherlands (J.L.B.); and Dept of Medicine, Newton-Wellesley Hosp, Newton, Mass (K.B.I.). Received Nov 14, 2002; revision requested Jan 20, 2003; final revision received Jun 2; accepted Jun 18. Portions of this work are sponsored by the U.S. Dept of the Army under DAMD 17-99-2-9001. Address correspondence to G.S.G.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The authors developed a decision model to compare the costs and effectiveness of UAE and hysterectomy. In the model, a cohort of women aged 40 years with a diagnosis of uterine fibroids and no desire for future pregnancy was followed up until menopause. The analysis was performed from a societal perspective, including all costs and effects, regardless of who incurs them. Transition probability and quality-of-life estimates were obtained from the literature and a gynecologist, whereas costs (in 1999 U.S. dollars) were estimated by using rates of Medicare reimbursement for hospital costs and physician fees. Sensitivity analyses of key estimates were performed. Results were expressed in costs per quality-adjusted life-year (QALY).
RESULTS: UAE was more effective (8.29 vs 8.18 QALYs) and less expensive ($6,916 vs $7,847) than hysterectomy. Cost-effectiveness results, with the exception of quality-of-life data, were robust to changes in most model assumptions. When the quality-of-life adjustment was eliminated, the two procedures were equally effective.
CONCLUSION: UAE is a cost-effective alternative to hysterectomy across a wide range of assumptions about the costs and effectiveness of the two procedures. However, the study results were sensitive to changes in quality-of-life values.
© RSNA, 2004
Index terms: Arteries, therapeutic embolization, 854.1264, 854.1266, 98.1264, 98.1266 • Cost-effectiveness • Uterine neoplasms, 854.315, 854.318 • Uterine neoplasms, therapy, 854.1264, 854.1266, 98.1264, 98.1266
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Symptomatic fibroids have generally been treated with hysterectomy. This common surgery is invasive, however: Complications occur with approximately 10%–15% of surgeries, and several days of hospital stay and a recovery time of about 6 weeks can be expected (2,3). In 1998, approximately 645,000 hysterectomies were performed in the United States, making this surgery second to only cesarean section as the most common gynecologic surgical procedure (4). More than one-third of hysterectomies are performed for the treatment of uterine fibroids, making these tumors the largest single indication for the surgery (5).
Concerns about quality of life and cost containment have prompted the development of several treatment alternatives to hysterectomy. A promising minimally invasive treatment, uterine artery embolization (UAE), has emerged in the past few years. To perform UAE, an interventional radiologist inserts embolic material into the branches of the uterine artery, blocking blood flow to the fibroid and causing it to shrink (6). The procedure has so far been shown to be safe and effective in the short term, with low complication rates andreduced hospital stay lengths and recovery times (7). However, the potential viability of UAE as an alternative to hysterectomy also depends on many factors in the long term, such as fibroid recurrence; quality of life; need for repeat procedures; and costs to the patient, her employer(s), and the health care system. To our knowledge, there have been no published studies to analyze the cost-effectiveness of UAE. Thus, the purpose of our study was to compare the cost-effectiveness of UAE with that of hysterectomy for women with symptomatic uterine fibroids.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
In the hysterectomy strategy, patients could either die as a result or have a major complication of the procedure or have neither of these outcomes. As a result of hysterectomy, the surviving patients would be either cured or not cured. A certain percentage of patients treated for fibroids do not benefit from hysterectomy, because even though their symptoms are initially thought to be caused by fibroids, they are found to be "nonspecific pelvic pain" (3). We assumed that neither the cured nor the noncured patients underwent further interventions after hysterectomy. Any additional interventions for unrelieved symptoms were not explicitly modeled.
We used "no treatment" as the reference strategy in our analysis. In this strategy, patients incurred condition-related costs and had rates of age- and sex-specific death based on standard 1998 U.S. life tables (9).
In our base-case analysis, we included 40-year-old premenopausal women with a diagnosis of symptomatic uterine fibroids. Because the effect of UAE on fertility is unknown, the results of our analysis should be applied only to women with no desire for future pregnancy. In the analysis, we chose to evaluate the quality-adjusted life expectancy and costs until menopause, which occurs, on average, at age 51 (10). After menopause, we expected no clinical or cost difference between UAE and hysterectomy. The effects of UAE or hysterectomy on future related events, such as heart disease and cancer, are controversial and were beyond the scope of this analysis.
In performing this analysis, we followed the recommendations of the U.S. Panel on Cost-Effectiveness in Health and Medicine (11). As recommended by this panel, we estimated effectiveness and costs from a societal perspective, including all effects and costs, irrespective of who incurs them. We calculated incremental cost-effectiveness ratios (ICERs), with differences in costs between the two strategies included in the numerator and differences in additional health benefits between the two strategies included in the denominator. All survival rates and costs were discounted at an annual rate of 3%. All analyses were performed by using computer software programs (DATA 3.5, TreeAge, Williamstown, Mass; and Microsoft Excel 97, Microsoft, Redmond, Wash).
Data Sources and Assumptions: Effectiveness
Parameter estimates of treatment effectiveness in the model were obtained from published literature and a gynecologist (J.B.I.). In the absence of randomized controlled trials, we gave priority to studies with larger sample sizes and longer follow-up times (M.T.B.). Table 1 shows the estimates and sources used for our base-case analysis and the ranges used for the sensitivity analyses.
|
Long-term results.—Of particular concern is the long-term performance of UAE—specifically, whether it is possible for the fibroid to recur after a successful UAE procedure. To our knowledge, no such recurrence has been reported in the published literature so far; therefore, in our base case, we assumed that there was no chance of recurrence. However, in subsequently performed sensitivity analyses, we assumed that the recurrence rate would not exceed 1% annually. Long-term life expectancy was calculated by using age- and sex-specific mortality rates from standard 1998 U.S. life tables (9).
Health-related quality of life.—We assumed that the patients who were not cured with UAE or hysterectomy had a postoperative quality of life similar to that before the treatment, when menorrhagic. Although the patients who were not cured after hysterectomy no longer had menorrhagia, we assumed that the effect of their symptoms on their quality of life was similar to the effect of the menorrhagia. Because utilities for the cured state after UAE were not available from the literature, we assumed that these were similar to those of women in the same age group in the general population. Table 2 shows quality-of-life weights for the base case.
|
Data Sources and Assumptions: Costs
Costs were computed from a societal perspective—that is, not only hospital and physician costs but also patient and follow-up costs were included. Costs were computed from the 1999 Medicare Provider Analysis and Review database. Additional cost information was retrieved (M.T.B.) from our institution’s hospital accounting database (Transition Systems, subsidiary of Eclypsis, Delray Beach, Fla). The system calculates costs by using relative value units to represent both the technical difficulty costs and the time costs associated with the procedures. Costs derived from this system are generally considered to more accurately reflect resource use than charges, even when converted to costs by using cost-to-charge ratios (15,18). All costs were converted to 1999 U.S. dollars by using the medical component of the consumer price index (19). Table 2 shows cost estimates for the base case.
Hospital and physician costs.—We computed the cost of the UAE procedure on the basis of Medicare reimbursements (ie, metropolitan Boston rates) for American Medical Association–designated Common Procedural Terminology (CPT) codes by using the typical coding scenario from the Society of Interventional Radiology (20,21). This scenario includes the following codes: 75736*2, 36247*2, 37204, 75894, and 75898*2. With the multiple surgical discount incorporated, the total for hospital and physician costs was $4,470.
At our hospital, patients who undergo UAE for fibroids require, on average, a single day of observation after the procedure. According to our hospital’s cost accounting system, the hospital cost for observation of these patients was $997 in 1999. Therefore, the total procedural cost for UAE was estimated to be $5,467. We also assumed that 6 months after the UAE procedure, patients would require a follow-up visit and magnetic resonance (MR) imaging. The costs for these services were estimated by adding the reimbursements for CPT code 99213 for the visit and CPT code 72196 for pelvic MR imaging and the cost of one-half day of lost wages, for a cost of $668.
Hospital costs for hysterectomy were based on the Medicare reimbursements corresponding to the appropriate diagnosis-related groups (22). We computed a weighted average for diagnosis-related groups 358 and 359 on the basis of the distribution of a cohort of patients who underwent hysterectomy for fibroids at our institution between 1998 and 2001. We approximated that the average hospital cost for hysterectomy was $3,799. Physician costs for hysterectomy were computed on the basis of reimbursement for CPT code 58150 (ie, total abdominal hysterectomy), for a total procedural cost of $4,795.
Patient costs.—The patient costs included in the model were time costs. Patient time costs were determined by multiplying the median weekly wage rate by the number of weeks spent recovering from the procedure and the productivity or leisure time, in weeks, lost in follow-up (23). Mean recovery time was assumed to be 1 week for UAE (7) and 6 weeks for hysterectomy (3). Patients who were not cured after treatment were assumed to have had the same condition-related costs that they had before treatment. Condition-related costs were estimated on the basis of activity loss reported in the National Health Interview Survey (24).
Morbidity and mortality costs.—The additional costs incurred owing to complications of the UAE or hysterectomy procedure—both major and minor—were included. The most important major complication of UAE that has been reported is pulmonary embolus (7). On the basis of that report, we assumed that this complication requires 4 days of observation and anticoagulation therapy. The costs associated with pulmonary embolus were estimated by adding the hospital, imaging, and patient time costs. Diagnosis-related group 078 was used for hospital costs, and CPT code 78588 was used for imaging costs. Lost leisure and work time was assumed to be 1 week in addition to the week lost due to the UAE procedure, for an additional cost of $473. Therefore, the total cost associated with the complication of pulmonary embolism was assumed to be $6,910. The minor complication for UAE included was that of an office visit for dilation and curettage (CPT code 58120) and a loss of 1 day of work or leisure time, for a cost of $339.
The major complication of hysterectomy included was bowel injury causing obstruction and requiring an additional day of inpatient observation. We assumed that the cost of bowel obstruction, or that of any other major complication incurred as a result of the hysterectomy, would be included in the average Medicare reimbursement for hysterectomy with complications (diagnosis-related group 358). However, we also added the costs incurred owing to an increased length of hospital stay and lost patient time, making the total cost for this complication $1,654. We assumed that the cost of any minor complications of hysterectomy, such as hemorrhage requiring transfusion, wound infection, or urinary tract infection, would be included in the average Medicare reimbursement. In addition, these complications do not require that the patient stay in the hospital longer than the typical range of lengths of stay for hysterectomy. The costs associated with procedure-related deaths (within 30 days after the procedure) were estimated to equal 150% of the procedural cost.
Sensitivity Analyses
Sensitivity analyses of model assumptions, uncertain model transition probabilities, and costs (Tables 1 and 2) were performed (M.T.B.).
We analyzed the effects of uncertainty about the short-term effectiveness of both UAE and hysterectomy—specifically, in terms of procedure-related mortality rates, complication rates, recovery time, long-term outcomes, and secondary treatment options (ie, no further treatment for women not cured with UAE). For the sensitivity analyses, the recovery times for hysterectomy were reduced to 4 and 2 weeks to reflect the use of less invasive surgical alternatives to the abdominal approach, such as vaginal or laparoscopically assisted hysterectomy. In addition, we evaluated the effect of quality of life as well as no quality adjustment (ie, utilities of 1.0 for all health states).
There is much uncertainty regarding the actual costs associated with either UAE or hysterectomy. In our base-case analysis, we estimated costs on the basis of Medicare reimbursement rates. In subsequent sensitivity analyses, we used hospital costs retrieved from our institution’s accounting system (25). In our base-case analysis, we assumed that the cost of procedure-related mortality was 150% of the cost of the procedure itself. However, with this gross estimation, the costs of death related to one procedure would be different from the costs of death related to the other procedure, which is an implausible scenario. To rectify this, in sensitivity analyses, we set this cost as equal in the two procedures by calculating the mean of 150% of the procedural costs. We also increased the mortality cost to 200% of the cost of each procedure. Additional sensitivity analysis data were computed by reducing the mean age of the patient cohort to 30 years, varying the mean patient age at onset of menopause from 46 years to 56 years, and varying the annual discount rate between 0% and 5%.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
The cost-effectiveness results were also sensitive to changes in the utilities for menorrhagia and changes in the effects of treatment. For example, when the utility for menorrhagia was increased to 0.90, UAE was still more effective than no treatment and hysterectomy; however, the ICER for UAE versus no treatment increased to $15,285 per QALY. With elimination of the quality-of-life adjustment altogether, UAE and hysterectomy were essentially equally effective.
In a two-way sensitivity analysis, we evaluated how changes in posttreatment quality of life affected the cost-effectiveness results (Fig 2). Overall, if the utility of cured after UAE was higher than the utility of cured after hysterectomy, then UAE dominated hysterectomy. However, when the utility of cured after UAE was reduced from 0.90 to 0.85, hysterectomy was more effective and had an ICER that did not exceed $75,000 per QALY.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In general, the results were not sensitive to changes in model parameters such as procedure-related mortality, cure rates, and complication rates. The most notable exceptions were costs and quality of life. Increasing the procedural costs of UAE made it more expensive than hysterectomy, in spite of the savings in recovery time. Because both UAE and hysterectomy are considered safe procedures and are associated with very low mortality risk, neither will confer a survival benefit. Therefore, any difference in effectiveness between the two procedures will hinge on the effect that they have on quality of life. A great amount of uncertainty exists in this area, however. For example, the quality-of-life effect of fibroids, represented in our model as menorrhagia, is unknown. The effect of UAE in particular on quality of life also is uncertain.
Another area of uncertainty is the quality of life during the recovery period after either UAE or hysterectomy. Our choice of a 20% reduction in quality of life was somewhat arbitrary, given the lack of available data. However, because in previously published cost-effectiveness analyses (26–28) of much more aggressive surgeries, the quality of life was reduced by 30% during the recovery period, we believed that our choice of reduction value was justified. Because our cost-effectiveness results were sensitive to changes in quality-of-life values, it is clear that this research area needs to be addressed.
Our study had several limitations. By necessity, our model represents a simplification of clinical practice in reality. In particular, we did not explicitly model any interventions after hysterectomy. In addition, hysterectomy or no treatment was the only option for women who were not cured with UAE. In reality, these patients may have a host of alternatives, such as myomectomy or medical management. Furthermore, ovarian failure, which is an important complication associated with UAE, was not included in our model. The risks and costs associated with ovarian failure remain controversial; however, not including them in the model may have led to an overestimation of the effectiveness of UAE.
Another simplifying assumption was made regarding the use of MR imaging. Only the costs of MR imaging performed after UAE were included in this model. At some institutions, however, MR imaging is performed before and after UAE, and at others, no MR imaging examination is performed in association with UAE. The costs of MR imaging are substantial, so this assumption may have caused an over- or underestimation of the costs of UAE, depending on institutional standards of care.
In this study, we compared UAE with hysterectomy and no treatment as if all patients would elect to undergo one of the two procedures. Many women are very concerned about preserving their fertility, and their decision-making processes are greatly influenced by this concern. Because of this factor, the results of this study should be applied only to women with no desire for future pregnancy. However, if UAE is found to have no effect on fertility, there may be additional quality-of-life gains associated with UAE that were not explored in this study.
Another limitation of our study is related to the uncertainty surrounding several model parameters. We attempted to include reasonable parameter estimates based on expert opinions from the literature; however, we acknowledge that there is uncertainty as to the exact values of these parameters. Our approach, therefore, was to vary these values within plausible ranges by means of sensitivity analyses.
In conclusion, our study results suggest that UAE is a cost-effective alternative to hysterectomy for the treatment of women with symptomatic uterine fibroids. These results were robust to changes in many assumptions about the effectiveness and costs of the procedures. However, the results were sensitive to changes in the effects of fibroids and the treatment option on quality of life.
|
FOOTNOTES |
|---|
The information presented herein does not necessarily represent the position of the government, and no official endorsement should be inferred.
Author contributions: Guarantors of integrity of entire study, M.T.B., J.L.B., G.S.G.; study concepts, all authors; study design, M.T.B., J.L.B., G.S.G.; literature research, M.T.B.; data acquisition, M.T.B.; data analysis/interpretation, M.T.B., J.L.B., G.S.G.; statistical analysis, M.T.B.; manuscript preparation and definition of intellectual content, M.T.B.; manuscript editing, M.T.B., J.L.B., G.S.G.; manuscript revision/review and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Tropeano, S. Amoroso, and G. Scambia Non-surgical management of uterine fibroids Hum. Reprod. Update, May 1, 2008; 14(3): 259 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Bucek, S. Puchner, and J. Lammer Mid- and Long-Term Quality-of-Life Assessment in Patients Undergoing Uterine Fibroid Embolization. Am. J. Roentgenol., March 1, 2006; 186(3): 877 - 882. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
