10,000
9,000
8,000
7,000
6,000
5,000
4,000
3,000
2,000
1,000
0
1998 1999 2000 2001 2002 2003 2004 2005
The analysis shows that
on average, bariatric
surgery costs were fully
recovered after 53 months
for procedures taken as
a whole. Costs of open
surgery performed between
1999 and 2002 were fully
recovered after 77 months,
and costs of open surgery
performed between 2003
and 2005 were recovered
after only 49 months.
fully recovered after 77 months, and costs of open surgery performed between 2003 and 2005 were recovered after only 49
months. The study’s authors attribute the improved return
on investment to greater surgical experience, enhanced technologies, and still-developing Centers of Excellence, where
improved outcomes and follow-up contribute to lower costs.
For laparoscopic bariatric procedures, which constitute
the predominant surgical approach today, costs were fully
recovered much sooner—within 25 months. Current weight-reduction surgery involves a variation on one of three types
of procedures: gastric bypass, gastric band, and sleeve gastrectomy. When performed as a laparoscopic rather than open
surgery, the procedure is less invasive and less costly, which,
therefore, allows faster recovery of initial costs.
The Actuarial Test
Using the findings of the 2008 study, it’s possible to argue that
covering bariatric surgery represents an effective cost-management strategy for health plans. The findings certainly merit
further examination under an actuarial lens.
Is the Study’s Claim Experience Relevant?
The study is based on a data set, incorporating claims experience, for 5,472,542 lives covered under employment-based
health plans. These data included both medical and prescription drug claim costs. The claim data set, therefore, should be
indicative of what a health plan would experience in an employed population.
Insurers providing coverage for bariatric surgery, in fact,
may be able to replicate this study using their own claims data.
Health plans that aren’t offering coverage can use the study as
a benchmark until they develop their own data.
Is Morbid Obesity Prevalent and Cost Significant?
The study findings confirm that the prevalence and cost of morbid obesity are substantive and warrant further consideration
by health plans.
In the study, 36,384 covered lives had at least one morbid
obesity diagnosis, for a prevalence rate of 6. 6 per 1,000 ( 1,000
x 36,384 / 5,472,542). We also know that claims for morbid
obesity typically are underreported because health care providers generally are paid for treating the associated conditions
(diabetes, hypertension, etc.) rather than the obesity itself. It’s
likely that the actual prevalence of morbid obesity in the study
population was higher than 6. 6 per 1,000.
When we consider the cost of morbid obesity, the study
paints a compelling picture. Based on data used in the study,
Chart 2 plots the higher average annual health care costs for
the morbidly obese patient compared with the average patient.
Note that costs associated with morbid obesity also generally
increased more quickly, except in the last year of the study.
The study also revealed that a significant number of morbidly obese patients receive care that seems to address vague
symptoms rather than underlying causes. Table 1 illustrates
the percentage of morbidly obese patients present in 2005
who received care for selected diagnoses included in the signs,
symptoms, and ill-defined conditions group.
Claims for ill-defined conditions appear to be quite common
among the morbidly obese. The more imprecise the diagnosis,
the more likely these claims represent spending by health plans
that provide limited value to the patient.
Is the Study Reliable?
The research methodology and design employed by Crémieux
and his fellow researchers differ from prior studies of bariatric surgery in several ways that ensure a reliable foundation for
actuarial use.
■ ■ The study determined cost and savings by comparing bariatric surgery patients with control group patients who didn’t
have the surgery. Surgery patients were matched to nonsurgery
(control group) patients on the basis of relevant data (
demographics, diagnoses, pre-surgery cost, etc.). Diagnostics on the
baseline characteristics of the surgery patients and their controls revealed no systematic differences between the groups that
