Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Mucosa
 The
 Esophagogastric  Junction
 Barrett's
 Esophagus

  Browse by Author
  Browse by Movies
OESO©2015
 
Volume: Barrett's Esophagus
Chapter: Screening and surveillance
 

What information can the computer cohort simulation of Barrett's patients contribute to cost-effectiveness analysis of cancer risk?

D. Provenzale (Durham)

There is an increased risk of adenocarcinoma in patients with Barrett's esophagus. The risk is estimated at 30-125 times that of the general population. Patients with high-grade dysplasia (HGD) are at greatest risk for developing adenocarcinoma. Because of their cancer risk, current recommendations are for periodic endoscopic surveillance to detect dysplasia and early cancer. Most authorities recommend esophagectomy for HGD, but because of the morbidity and mortality associated with esophagectomy, others reserve this procedure only for those with cancer. Previous reports have suggested that the incidence of cancer in patients with Barrett's ranges from 0.2% annually or 1 in 441 patient years to 2.1% annually or 1 in 48 patient years. Most recent reports, however, suggest that the incidence of cancer is in the range of 0.4% to 0.5% annually or 1 in 251 to 1 in 208 patient years. Our question for this analysis was, is surveillance of patients with Barrett's esophagus cost-effective compared to common medical practices? Because there were no controlled trials that examined the efficacy of surveillance, we created a computer cohort simulation of 50 year old patients with Barrett's esophagus and no evidence of dysplasia by biopsy. The strategies we included in our model are shown in Table I. We examined surveillance every 1-5 years and no surveillance in which endoscopy was performed for new or worsened dysphagia. For each of the strategies on the left, the indication for surgery that we considered was HGD. From published literature, the parameters we included in our model are shown in Table II. From published reports we used complications of an endoscopy of 1.3 procedures and an average incidence of cancer of 0.4% annually or 1 in 227 patient years. We used our measured value for quality of life after esophagectomy. On a scale from 0 to 1, with 0 equivalent to being dead, and 1 equivalent to a state of perfect health, we measured quality of life in patients who had undergone esophagectomy for HGD or cancer one year earlier at Duke University. On a scale from 0 to 1, they rated their quality of life at 0.97. Although they were symptomatic after esophagectomy, they believed that they had been cured of their cancer and this was extremely important to them [1].The criterion, or unit of measurement for our analysis was the incremental cost-utility ratio. This is the change in cost divided by the change in quality adjustedlife expectancy to move from no surveillance to surveillance or to move from less frequent surveillance to more frequent surveil-lance. The results are reported as dollars per quality-adjusted life year gained. We employed the standard technique of discounting in our analysis, which considers that a dollar today is worth more than a dollar in the future, and because life years are valued relative to dollars in these economic analyses, they are also discounted. The costs we included in our model are shown in Table III. These are direct medical costs from Duke Uni-versity. They are the actual resource costs for providing these services. This is in distinction to charges which may be falsely inflated or deflated based on profit or loss to the institution. These are facility costs and medicare reimbursement for

Table I. Strategies.

Table II. Parameters.

Table III.

physician fees. The cost of endoscopy is $ 600. Endoscopy with complication is $ 3,700. The costs for esophagectomy and perforation requiring surgery ranging from $ 7,000 to $ 23,800. The cost for annual follow-up post esophagectomy is $ 1,000 and the cost for the care of the terminally ill patient with esophageal cancer at $ 33,900 [1].

Our results are shown in Figure 1. The vertical axis displays the average life time cost per patient and the horizontal axis displays the discounted quality-adjusted life expectancy in years. Each of the circles represents the results of an alternative management strategy. On the bottom left, no surveillance costs approximately $ 3,000 and is associated with living approximately 12.65 additional years after entering a surveillance program. Surveillance every 5 years costs approximately $ 12,500 and is associated with living an additional 12.74 quality-adjusted life years after entering a surveillance program. The change in cost divided by the change in life quality-adjusted expectancy to move from no surveillance to surveillance every 5 years is the incremental cost utility ratio shown on the top of the line at $ 98,000 per quality-adjusted life year gained. Focusing on the more frequent surveillance intervals every 4, 3, 2 and 1 year, the line does not connect these strategies. That is, because they are inferior to surveillance every 5 years because they cost more and yield a lower life expectancy. Costs are higher and life expectancy is lower because on average, there are more endoscopies, endoscopic complications and surgeries in these strategies. Now turn your attention to the text at the top of the slide. Compared to published incremental costeffectiveness ratios our results show that surveillance every 5 years has an incremental costutility ratio of $ 98,000 per quality-adjusted life year gained. That falls somewhere in between the costs of breast cancer screening [2] and heart transplantation [3]. So, surveillance every 5 years would be considered cost-effective when compared to these other well accepted medical practices. In order to examine the cost-effectiveness of surveillance given a limited health care budget, we performed a sensitivity analysis on the willingness to pay or budget of those who make health policy and on the incidence of cancer, a critical parameter in decisions for surveillance and esophagectomy in patients with Barrett's esophagus (Figure 2). The vertical axis displays different levels of willingness to pay or budgets of those who make health policy, ranging from $ 25,000 to $ 300,000 per qualityadjusted life year gained. The horizontal axis displays the incidence of cancer ranging

Figure 1. Results.

Figure 2. Sensitivity analysis-incidence of cancer.

from 0% to 4%. The baseline is at 0.4%. There are no budgetary constraints. The lower bar shows that surveillance every 5 years provides the greatest gain in quality-adjusted life expectancy. Moving up to the next horizontal bar of Figure 2, for the policy maker with $ 300,000 to spend on surveillance, similar to the cost of cervical cancer screening showed earlier in Figure 1, it is shown that surveillance every 5 years is the only strategy that both increased life expectancy and would be within the budget of the policy maker with $ 300,000 to spend. For the policy maker with only $ 100,000 to spend on surveillance, similar to the cost of heart transplantation, surveillance every 5 years would increase life expectancy and would be within the budget of the policy maker with $ 100,000 to spend. For the policy maker with only $ 50,000 or $ 25,000 to spend on surveillance, similar to the cost of mammography for breast cancer screening, it appears that with this limited health care budget, no surveillance is the only strategy that would not exceed the $ 50,000 or $ 25,000 limit.

Conclusions

o Surveillance of patients with Barrett's esophagus is effective.

o Surveillance every 5 years provides the greatest gain in quality adjusted life expectancy and has an incremental cost-utility ratio similar to those of common medical practices.

o The cost-effectiveness of surveillance is based on the budget of those who make health policy.

o Future research on the incidence of cancer and the accuracy of surveillance endoscopy are crucial for determining the most effective and cost-effective surveillance interval.

References

1. Provenzale D, Schmitt C, Wong JB. Barrett's esophagus: a new look at surveillance based on emerging estimates of cancer risk. Am J Gastroenterol 1999;94:2043-2053.

2. van der Maas PH, de Konig JG, van Ineveld BM, et al. The cost-effectiveness of breast cancer screening. Int J Cancer 1989;43:1055-1060.

3. Pennock JL, Oyer PR, Reitz BA, et al. Cardiac transplantation in perspective for the future:survival complications, rehabilitation, and cost. J Thorac Cardiovasc Surg 1982;83:168-177.


Publication date: August 2003 OESO©2015