cost-effectiveness of sibutramine in the lose weight study

458 Journal of Managed Care Pharmacy JMCP Vol. 11, No. 6 July/August 2005 www.amcp.org

n 2000 almost 55% of adults in the United States aged 20 years or older were overweight (body mass index [BMI]of ≥ 25 kg/m2).1 More recent data suggest that, in 2003, all

states in the United States had greater than 15% of their populations who were considered obese (BMI of > 30 kg/m2),with 4 states (Alabama, Indiana, Mississippi, and West Virginia)reporting that more than 25% of their residents were obese.2

The incidence of chronic diseases, including hypertension,hyperlipidemia, type 2 diabetes, coronary artery disease, stroke,gallbladder disease, musculoskeletal disorders, and some cancers,increases with increasing weight.2,3 Direct health care costs forchronic diseases related to obesity were estimated at $47.5 billionin 2002 dollars.4 When indirect costs associated with obesityare considered, the estimated total cost of obesity in the UnitedStates was more than $99 billion in 1995 dollars.5 Eighty-fivepercent of the costs of obesity are considered attributable to 5 diseases: hypertension, type 2 diabetes, coronary artery disease, hyperlipidemia, and stroke.2,6

The relationship between obesity and other chronic conditionshas been well studied. Persons with BMI values of 25 kg/m2 areat increased risk of hypertension, type 2 diabetes, coronaryheart disease, stroke, gallbladder disease, osteoarthritis, sleep apnea,

ABSTRACT

OBJECTIVE: No previous studies conducted in the United States have evaluatedthe cost-effectiveness of drug therapy when used in conjunction with a weightmanagement program (WMP) for treatment of obesity. The objective was to compare the cost-effectiveness of sibutramine plus a structured WMP versusonly a structured WMP in both overweight and obese individuals. The core WMPwas a physician-supervised, multidisciplinary program for which each enrolleepaid $100 out of pocket.

METHODS: A cost-effectiveness analysis was performed based upon the resultsof a previously published randomized controlled trial conducted within a managed care organization. The target population for this study was obese oroverweight persons. The perspective of the study was that of a managed careorganization. The intervention consisted of subjects receiving a WMP with or without sibutramine. The primary outcomes of this study were (a) absolutechange in body weight and percentage change in body weight over 12 months,(b) change in obesity-related and total medical costs from 12 months prior toenrollment through 12 months after enrollment, and (c) cost-effectiveness interms of cost per pound of weight loss. All costs were adjusted to 2004 dollarsusing the respective components of the consumer price index for each medicalservice or medication.

RESULTS: A total of 501 evaluable subjects were enrolled in the study, with 281 receiving sibutramine plus a structured WMP and 220 receiving only thestructured WMP. The mean±SD weight loss was significantly greater in the sibutramine (13.7±15.5 pounds, 4.8%) group than in the nondrug group (5±13.2pounds, 2.2%) (P < 0.001). The change in obesity-related total cost was a medianincrease of $408 for the sibutramine group compared with $31 for the nondruggroup (P < 0.001). The change in total health care cost was a median $1,279increase in the sibutramine group compared with $271 for the nondrug group (P < 0.001). Adding sibutramine to the WMP increased the total cost by $44 peradditional pound of weight loss (95% confidence interval, 42-46). Sensitivityanalyses found that the results were sensitive to the price of sibutramine,whereas varying the cost of clinic visits did not substantially change the results.

CONCLUSION: Patients enrolled in a WMP receiving sibutramine had greaterweight loss and decrease in body mass index at greater cost than did patientsenrolled in the same program who did not receive sibutramine. There were no observed savings in total health care resource utilization or cost in the sibutramine group compared with the nondrug group.

KEYWORDS: Obesity, Sibutramine, Cost-effectiveness, Cost analysis

J Manag Care Pharm. 2005;11(6):458-68

I

DANIEL C. MALONE, PhD, is an associate professor, Colleges of Pharmacy andPublic Health, University of Arizona, Tucson; MARSHA A. RAEBEL, PharmD, is research manager, pharmacotherapy and clinical trials, Kaiser Permanente ofColorado (KPCO), Denver, and a clinical associate professor, School of Pharmacy,University of Colorado at Denver and Health Sciences Center; JULIE A. PORTER,PharmD, is a cardiovascular regional medical liaison, sanofi-aventis, Bridgewater,New Jersey (at the time of this study, she was a clinical pharmacy specialist, KPCO,Littleton); FRANCES A. LANTY, PharmD, is the drug information coordinator,KPCO, Aurora, and is also an assistant clinical professor, School of Pharmacy,University of Colorado at Denver and Health Sciences Center; DOUGLAS A.CONNER, PhD, is a biostatistician and senior research analyst, Clinical Research Unit, KPCO, Denver; ELIZABETH C. GAY, MA, is director, PreventionDepartment, KPCO, Denver; JOHN A. MERENICH, MD, is director, PopulationManagement, KPCO, Denver, and a clinical associate professor of medicine,University of Colorado at Denver and Health Sciences Center; ERIN A. VOGEL,PharmD, is a clinical pharmacy specialist in primary care, KPCO, Aurora.

AUTHOR CORRESPONDENCE: Daniel C. Malone, RPh, PhD, AssociateProfessor, University of Arizona College of Pharmacy, PO Box 210207, Tucson, AZ 85721. Tel: (520) 626-3532; Fax: (520) 626-7355; E-mail: [email protected]

Copyright© 2005, Academy of Managed Care Pharmacy. All rights reserved.

Authors

ORIGINAL RESEARCH

Cost-Effectiveness of Sibutramine in the LOSE Weight Study:Evaluating the Role of Pharmacologic Weight-Loss Therapy

Within a Weight Management ProgramDANIEL C. MALONE, PhD; MARSHA A. RAEBEL, PharmD; JULIE A. PORTER, PharmD; FRANCES A. LANTY, PharmD;

DOUGLAS A. CONNER, PhD; ELIZABETH C. GAY, MA; JOHN A. MERENICH, MD; and ERIN A. VOGEL, PharmD

Note: An editorial on the subject of this article, “‘U Can’t Touch This’With Pharmacotherapy Alone for Weight Loss or Smoking Cessation”appears on pages 516-20 of this issue.

respiratory problems, and various forms of cancer (endometrial,breast, prostate, and colon).7 There is also a variety of complicationsthat arise from obesity, including pregnancy complications, stressincontinence, depression, and menstrual irregularities.7 Numerousstudies have found that obese persons are more likely thannonobese persons to have hypertension.8-12 Data from theInternational Study of Sodium, Potassium, and Blood Pressure(INTERSALT) found that higher body weight was associated with a3 mm Hg increase in systolic and 2.3 mm Hg increase in diastolicblood pressures.11 This increase is associated with an increased riskof coronary heart disease of 12% and a 24% increase in the risk ofstroke. There is a clear relationship between obesity and coronaryheart disease.13-15

Several studies have shown that increased BMI is associatedwith a higher prevalence of metabolic syndrome. Data from theThird National Health and Nutrition Examination Survey(NHANES III) show a marked increase in the prevalence ofmetabolic syndrome as a function of BMI.16 In persons with aBMI greater than 30, the prevalence of metabolic syndrome wasalmost 50% in white and Mexican American women. Amongwhite and Mexican American men, the prevalence of metabolic syndrome was greater than 50% when BMI was 31 or greater.This same study also found that metabolic syndrome is morecommon in white and Mexican American persons as comparedwith black persons. The odds ratio of having metabolic syndrome was greater than 5 for overweight males and femalesas compared with normal-weight persons. These odds ratiossharply increased when BMI was greater than 30 (25.2 for menand 14 for women). There also is a strong association betweenobesity and diabetes.17-19 The risk of diabetes was found toincrease by 25% for each additional BMI unit over 22 kg/m2.17

A more recent study found that 56% of new type 2 diabetes inthe Health Professionals Follow-up Study could be attributableto a weight gain of greater than 7 kg.19

The relationship between obesity and the myriad health conse-quences challenges managed care organizations. Most peoplebelieve that obesity is due more to caloric intake than genetic anddisease-mediated factors, despite evidence to the contrary.20 With anincreasing population of obese members, managed care organiza-tions are encountering increased pressure to offer solutions to theproblem, including the decision as to whether antiobesity drugsshould be included in the prescription drug benefit.

Because strategies to effectively lower body weight are needed,we conducted a study at Kaiser Permanente of Colorado(KPCO) to evaluate the effectiveness of obesity drug therapy:the Long-term Outcomes of Sibutramine Effectiveness onWeight (LOSE Weight) study.21,22 This prospective, randomized12-month study assessed the impact of sibutramine (Meridia)in combination with the Kaiser Permanente WeightManagement Program (WMP) compared with the WMP alone.The results of the primary clinical end point of weight loss werethe focus of a previous publication.22 A secondary end point of

the LOSE Weight study was an evaluation of health careresource utilization and associated costs.

Results of health care resource utilization, costs, and thecost-effectiveness of sibutramine in the context of the LOSEWeight study are described here. In this current study, we evaluated the cost-effectiveness of sibutramine plus the structured WMP versus only the structured WMP. We hypothesizedthat there were no differences in resource use and costs betweenthe group of patients who received sibutramine compared withthe nonsibutramine group and that the incremental cost-effectiveness would be no different from zero.

nn MethodsStudy Design, Setting, and PopulationThe LOSE Weight study was conducted at KPCO, a group-model health maintenance organization (HMO). In the year2000, KPCO provided health care for a diverse population ofmore than 350,000 members in the Denver-Boulder-Longmontmetropolitan area; about 50,000 of these individuals wereMedicare beneficiaries. At the time of this study, more than 90%of members were enrolled in a prepaid, traditional HMO insuranceplan; benefits did not differ between Medicare and non-Medicare enrollees. About 96% of health plan members hadprescription drug benefits that did not include coverage of medications used to treat obesity.

The LOSE Weight study was a prospective, randomized trialinvolving overweight and obese patients enrolled in the WMP.The methods and clinical outcomes of the LOSE Weight studyare described in detail elsewhere.22 Briefly, study enrollmentoccurred from January 1999 through June 2000. Eligible patientswere aged 18 years or older and had either BMI ≥30 kg/m2 orBMI of 27-29.9 kg/m2 with one or more comorbidities, includingdiabetes, hypertension, and/or hyperlipidemia. Subjects wererandomized to receive sibutramine or no drug therapy, but allpatients participated in the WMP. Participating subjects wererandomly assigned to either the drug or nondrug group using acomputer-generated random numbers table. Study assignmentswere placed in envelopes that were opened upon completion ofthe baseline visit.

The core WMP was a physician-supervised, multidisciplinaryprogram for which enrollees paid $100. It included 5 monitoredcare visits with a prevention specialist and attendance at 2 ormore group-format weight management seminars. At WMPmonitored-care visits, weight, height, body fat, vital signs, andwaist and hip circumference values were obtained. Patient education and goal setting for diet, exercise, and other lifestylemodifications were provided by a prevention specialist.Monitoring for efficacy and safety of weight-loss medicationsoccurred, if applicable. All study subjects also participated in 2 education programs offered by the American HeartAssociation: Slim for Life ($70) and Active for Life ($70). Theseprograms each had 10 classes. Study subjects incurred a total of

Cost-Effectiveness of Sibutramine in the LOSE Weight Study: Evaluating the Role of Pharmacologic Weight-Loss Therapy Within a Weight Management Program

www.amcp.org Vol. 11, No. 6 July/August 2005 JMCP Journal of Managed Care Pharmacy 459



$240 in financial commitments to be included in the WMP andAmerican Heart Association programs, the same costs aspatients enrolled in these programs who were not study partic-ipants. Study participants who completed 6 months of thestudy received a $50 gift check. Those patients who completed12 months of the study received an additional $100 gift check.

Patients randomly assigned to sibutramine paid the usualpharmacy price for the drug at the time of dispensing and werelater reimbursed 75% of this price upon presenting the prescription receipt. Enrollment in the LOSE Weight study didnot influence the routine medical care these patients received.Study subjects continued to see primary care and specialtyphysicians at their discretion throughout the study. The KaiserFoundation Research Institute Institutional Review Boardapproved this study.

The primary measure of effectiveness for the clinical analysis wasthe mean change in body weight from baseline to 6 months afterenrollment; the secondary measure of clinical effectiveness was themean change in body weight from baseline to 12 months. For theeconomic analysis, we examined change in body weight through 12 months. The last observation carried forward was used for missing data. Because change in body weight can be influenced bythe initial starting weight, a secondary measure of effectiveness usedin this analysis was the percentage change in body weight.

Economic End PointsThe economic analysis in the LOSE Weight study measuredresource utilization over a 24-month period, including the 12 months before and after study enrollment. Utilization datawere derived from electronic medical records and administrativeclaim and clinical databases maintained and used by KPCO for providing patient care and tracking medical expenses. The analysis examined the use of medical resources from theperspective of a managed care organization. Evaluation ofhealth care resource utilization included outpatient visits(including medical office and emergency department visits),hospitalizations, professional services claims (e.g., oxygen,ambulance, outside physician referrals), and prescription medications. We have included total costs in the analysis to capture adverse events that may arise due to treatment withsibutramine. In addition, we also analyzed obesity-related utilization and costs.

Obesity-related and nonobesity-related health care resourceutilization was evaluated. Expert panels comprising 3 to 5physicians, 2 clinical pharmacists, and a health promotion specialist determined whether outpatient visits and hospitalizationswere related to obesity for the time periods of 12 months beforeand 12 months after study enrollment. Each evaluator conductedan independent review and was blinded to the identity of thetreatment arm. All experts were asked to independently classifymedication use as “1” (most likely related to obesity, points=1),“2” (possibly related to obesity, points=2), or “3” (not related to

obesity, points = 3). For a resource to be considered obesity-related, a majority of the panel members must have assigned avalue of 1, and no expert could assign a value of 3. Outpatientvisits were considered related to obesity if the panel determinedthat the visit was related to either obesity or to adverse eventsassociated with obesity therapy. Examples of conditions relatedto obesity included diabetes mellitus (type 2), pure hypercho-lesterolemia, hypertension, coronary arteriosclerosis, and gout. A hospitalization was considered related to obesity if one or more of the first 4 discharge diagnosis codes (International Classification of Diseases, Ninth Revision, Clinical Modification) on the hospital claimwere determined by the panel to be obesity-related.

To establish if medications were obesity-related, an expertpanel consisting of 3 physicians and clinical pharmacists wasasked to determine if the medications were obesity-related.Similar to medical encounters, medications were rated byexperts as “1” (most likely related to obesity, points = 1), “2”(possibly related to obesity, points = 2), or “3” (not related toobesity, points = 3). The same scoring rules applied to medica-tions when determining if a drug was used to treat an obesity-related condition. Examples of medications classified as beingobesity-related included antidiabetics, antihypertensives, dyslipidemic therapies, and nitroglycerin.

Resource Units and CostsCost estimates for outpatient visits and hospitalizations wereestimated using Medicare’s resource-based relative value scalefee schedule and the reimbursement rates of diagnosis-relatedgroups.23 The cost of each service was calculated by multiplyingthe number of units by a standard price per unit. Costs fromMedicare fee schedules were based on 1998 reimbursementrates. Costs for professional claims were based on billedamounts. The cost for prescription medications, including thestudy drug sibutramine, was based on average wholesale priceat the time of dispensing. All costs were adjusted to 2004 dollarsusing the respective components of the consumer price indexfor each medical service or medication.24

Economic AnalysisIncremental cost-effectiveness ratios were constructed using meanchange in total costs between the 2 groups divided by the meanchange in body weight. Incremental cost-effectiveness was also calculated on percentage change in body weight. Sensitivityanalyses were conducted using median measures of cost and effectbecause cost data were skewed. Comparisons also were madebetween groups with respect to obesity-related care. In addition,the cost of sibutramine was removed from the analysis to examinethe impact of sibutramine costs. Sensitivity analyses also were conducted by increasing clinic costs upward by 50%.

Statistical AnalysisThe analysis was conducted using an intention-to-treat population.


Evaluable subjects were defined as those subjects who retainedmembership in KPCO for at least 12 months, received at leastone dose of sibutramine (for the drug group), and had at leastone recorded weight 4 weeks after being randomly assigned tothe treatment groups. For those persons with fewer than 12 months of clinical data, the last observation was carried forward in the analyses. The study was powered to detect a 10% difference in change in body weight between the groupsusing a 2-tailed alpha of 0.05 and power of 75%. The estimatedsample size necessary was 285 subjects per group.

Univariate analyses were performed on clinical and economicvariables, and nonparametric statistical procedures were usedfor variables that were not normally distributed. Therefore,instead of reporting mean and standard deviations for skeweddata, we report median and 5th and 95th percentiles. For incre-mental cost-effectiveness ratios, a nonparametric bootstrap with1,000 replications was performed using Stata Version 8 (CollegeStation, TX). Determination of 95% confidence intervals weremade using the bias-corrected approach.25

Regression analysis was used to isolate the effect of the medications and cost of therapy while controlling for other factors. Variables in the regression analysis included age, gender, chronic diseases, BMI, and study treatment arm.Chronic diseases were determined using the chronic diseasescore (CDS) method of Clark et al.26 The regression equationused was: COST = A + B1AGE + B2 CDS + B3BMI +B4ARM +B5BMI*CDS + B6AGE*CDS + B7GENDER + ERROR where Arepresents the intercept, ARM represents study group assignment,COST was the dependent variable and represents the total medical care cost for the 12-month period after enrollment, andeach Bn represents the coefficient for each specified variable.This analysis was conducted using SAS REG procedure (SASVersion 9.0, SAS Institute, Cary, NC).

nn Results A total of 1,564 subjects were invited to participate in the study,with 976 (62%) not participating due to lack of interest. Figure1 displays the enrollment and completion rates for the study. Of the 588 patients enrolled in the study, 501 (85%) were eligiblefor evaluation based on the intent-to-treat analysis. Of these,220 subjects received the WMP-only and 281 received sibu-tramine plus WMP. Patient baseline demographic characteristicsare in Table 1. Of the evaluable patients, those assigned to the sibutramine plus WMP were significantly older and had higherBMI values than did those receiving only the WMP. This difference was likely related to the fact that persons who wereenrolled in the study and randomly assigned to the nondruggroup prematurely discontinued the study because they wereseeking the medication. The distribution of BMI is shown inFigure 2. There were 57 (26%) receiving only WMP and 94 (33%) receiving sibutramine plus WMP who had a baselineBMI greater than 40 (P=0.07). Six percent of the sample was aged

65 years or older, evenly distributed between the 2 groups (P=0.47). Figure 3 displays the body weight of the subjects by gender

over time for the 2 treatment groups. The mean (SD) weightloss at 12 months in pounds was 13.7 (15.) (range: -85 to +20)for sibutramine plus WMP and 5 (13.2) (range: -79 to + 20) forthe WMP only (P<0.001). The percentage change in weight was-6% (6.7) and -2.2% (5.5) for the drug and nondrug groups,respectively (P<0.001). Figure 4 shows that 80.9% of the sub-jects randomized to only the structured WMP had less than a5% weight loss, as compared with 52.7% for those randomly

LOSE Weight Study ProfileFIGURE 1

1,564 Screened

588 Randomized Into 2 Groups

296 Assigned to Weight Management

Program Plus Sibutramine

292 Assigned to Weight Management

Program Only

5 Adverse Events2 Lack of Efficacy

60 Lost to Follow-up16 Protocol Violation21 Withdrew Consent

0 Adverse Events0 Lack of Efficacy

80 Lost to Follow-up21 Protocol Violation44 Withdrew Consent

192 Completed Trial 147 Completed Trial

281 Included in Intention-to-Treat Analyses

15 Excluded From Intention- to-Treat Analyses*

220 Included in Intention-to-Treat Analyses

72 Excluded From Intention-To-Treat Analyses*

976 Did Not Meet Inclusion Criteria or Were Not Interested

* All evaluable patients were included in the intention-to-treat analyses. Evaluable subjects were defined as subjects who retained membership with Kaiser Permanente of Colorado for the 6- and 12-month follow-up periods, received at least one dose ofstudy medication (if randomized to sibutramine), and had a body weight recorded atleast 4 weeks after randomization.




assigned to sibutramine plus WMP. In contrast, 19.6% of the subjects receiving sibutramine had a weight loss ≥10%, comparedwith 10% of those receiving only the WMP.

Table 2 displays the health care utilization and correspondingcost for enrolled subjects 12 months before study enrollment.There were significant differences between the 2 groups withrespect to all physician visits and obesity-related physician visits.Persons randomly assigned to the sibutramine group had more physician visits (median = 10) than the nondrug group (median = 8). The only significant difference with respect tocosts was that persons in the sibutramine group had higher obesity-related physician visit costs (median=$166.46) than didthose in the nondrug arm (median =$82.42), (P<0.001).

Table 3 displays the health care utilization for the 12-monthperiod after enrollment in the study. This utilization reflects allhealth care received after study enrollment, regardless ofwhether the subject had follow-up visits for the study. Themajor differences in utilization between the 2 groups afterenrollment were a higher number of obesity-related physicianvisits, total prescriptions, and obesity-related prescriptions inthe sibutramine-plus-WMP group as compared with the WMP-only group. Health care costs were also significantly greater forthe sibutramine-plus-WMP group compared with the WMP-only group with respect to obesity-related physician visits, totalprescriptions, obesity-related prescriptions, total health careexpenditures, and obesity-related health care expenditures.Because these comparisons reflect differences between thegroups without controlling for baseline differences, it may notbe appropriate to attempt to interpret these findings. Rather, theprimary analysis focused on the magnitude of change from 12 months before enrollment to 12 months after enrollment.

The change in medical resource consumption and expendituresfrom before-study to after-study enrollment is shown in Table 4.There were no differences in overall health care resources usedbetween the groups in outpatient visits (P = 0.80), hospitaliza-tions (P = 0.45), or professional service claims (P = 0.70) overtime. The sibutramine-plus-WMP group had more prescriptionmedications than the WMP-only group (P<0.001). The medianchange in the number of prescriptions was 9 for the sibutramine-plus-WMP group and 2 for the WMP-only group (P < 0.001). This change in the number of prescriptions waslargely due to the prescriptions for sibutramine. More than 98%of the prescriptions for sibutramine were dispensed with a days’supply of 35 or fewer. Patient compliance with sibutramine wasreported elsewhere8; briefly, the mean compliance was 76%.The change in total health care expenditures was significantlyhigher in the sibutramine-plus-WMP group than in the WMP-only group: a median of $1,279 for those receiving sibutraminecompared with $271 for those in the nondrug group (P<0.001).

As noted above, an expert panel rated medical resource useinto 1 of 3 categories with respect to whether it was obesity-related. Table 4 displays the change in medical resources for

Demographics of the Sibutramine and Nondrug Groups

TABLE 1

Sibutramine Group Nondrug GroupItem (n=281) (n=220) P Value

Age, mean ± SD (years) 47.2±10.7 49.6±10.4 0.01*

Age, range (years) 19-79 23-72

Gender (% male) 21 15 0.09†

Weight, mean ± SD (lbs) 283.3±49.9 227.8±40.8 0.01‡

BMI, mean ± SD (kg/m2) 38.6±7.0 36.8±5.6 0.01*

BMI, range (kg/m2) 30.0-68.6 27.9-56.8

Body fat, mean ± SD (%) 40.7±6.1 39.8±6.0 0.10*

Waist circumference (cm) 109.7±17.3 106.4 ±14.0 0.03*

Chronic disease score, $929 ± $1,040 $1,154±$1,108 0.02‡mean ± SD ($)

* Wilcoxon rank sum test.† Chi square.‡ Student t test.BMI = body mass index.

Distribution of Body Mass Index at Enrollment by Treatment Group

FIGURE 2

<29.9 30 to 32 to 34 to 36 to 38 to 40 to 45 to ≥5031.9 33.9 35.9 37.9 39.9 44.9 49.9

n Weight Management Program Onlyn Sibutramine Plus Weight Management Program

Weight by Treatment Group and GenderFIGURE 3

n

n n

s

s s

s

u

uu u

z

z

zz

n

300

280

260

240

220

200

180

25%

20%

15%

10%

5%

0%

Baseline 6 Months 12 Months

Nondrug FemalesNondrug Males

Sibutramine Females Sibutramine Males

278

260 264

249248

258

227

223216

212

218

213Wei

ght

(Pou

nd

s)

Study Visit


obesity-related goods and services from the year before thestudy to 12 months after study enrollment. Again, resource usewas similar between the 2 groups except for prescription medications and change in total expenditures. The medianchange in the number of obesity-related prescriptions was 3 forthe sibutramine-plus-WMP group, compared with 0 for theWMP-only group (P <0.001). The group receiving sibutramineplus WMP also had a higher change in total expenditures thandid the WMP-only group (P < 0.001). Obesity-related costs represented 32% ($408/$1,279) of all health care expendituresfor the sibutramine-plus-WMP group versus 11% ($31/$271)for the WMP-only group, over this time frame.

The a priori cost-effectiveness analysis specified that obesi-ty-related costs were to be used in the primary analysis. Thedenominator for the cost-effectiveness analysis was eitherweight loss (in pounds) or percentage change in weight. Theaverage weight loss and percentage weight loss are shown inTable 5. The average cost-effectiveness of sibutramine plusWMP was $32 for each pound of weight loss. For the WMP-only group, the average cost-effectiveness was $12 for eachpound of weight loss. The incremental cost-effectiveness ratio(ICER) was $44 per additional pound of weight loss (95% confidence interval [CI], $42-$46). When median costsand weight-loss values were used, the ICER was $42 per additional pound of weight loss. When the cost-effectivenessanalysis was conducted using percentage change in weight, theICER for sibutramine plus WMP was $101 per each additionalpercentage change in weight loss (95% CI, $99-$102).

A sensitivity analysis was conducted using total costs insteadof obesity-related costs only. The ICER for sibutramine plusWMP was $194 (95% CI, $188-$200) per additional pound ofweight loss when using mean values for cost and weight. Forthe additional percentage change in body weight, the ICER was$399 (95% CI, $391-$406) when total costs were used. Whenthe costs of sibutramine were excluded, the mean obesity-relatedtotal costs were $19 (SD $721) for the drug and $54 (SD $582) for the nondrug groups, respectively (data notshown in tables). In this situation, sibutramine plus the WMPwas the dominant strategy, with lower costs and a greater effect.A sensitivity analysis that increased the cost of clinic visits by50% had a negligible effect, changing the ICER from $44 to $45per pound of weight loss.

In a multivariate analysis (Table 6), the strongest predictor oftotal health care costs was the study arm, with those subjects receiving sibutramine plus WMP having an estimated annualcost of $474 more than subjects not receiving sibutramine(P < 0.001). Interaction terms were evaluated in the model butwere not significant.

nn DiscussionThe results of this study suggest that the cost-effectiveness ofsibutramine plus a structured WMP was approximately $44 per

additional pound of weight loss as compared with the structuredWMP alone. For this cohort of subjects, obesity-related costsrepresented a modest proportion of their total health careexpenditures, 32% ($408/$1,279) for the sibutramine-plus-WMP group versus 11% ($31/$271) for the WMP-only group.When all health care costs were included, the incremental cost-effectiveness of sibutramine plus WMP was $194 per additional pound ($423 per additional kilogram) of weight loss.This analysis examines weight loss over a 1-year period, but thebenefits of weight loss are likely to extend beyond such a timeperiod. Although a recent systematic review documents theclinical effectiveness of sibutramine,27 long-term studies of sibu-tramine (5 years or more) are needed to estimate the extendedcost-effectiveness.

The results of this study may be generalizable to other settings where a WMP exists and patients are responsible forpaying for part of their antiobesity medications. In this analysis,we use a nonparametric bootstrap approach to estimate thecost-effectiveness and corresponding CIs. The method we usedresampled the original data 1,000 times, in essence repeatingthe study 1,000 times. This allows us to obtain 95% CIs for thestudy. Therefore, although the trial was just 1 study, we have, ineffect, used a very large study (more than 200 subjects in eacharm) to simulate the cost-effectiveness 1,000 times.

Other pharmacological therapies for weight loss exist, butfew economic studies have been conducted evaluating them.Bharmal et al. evaluated the cost-efficacy of sibutramine andorlistat with a decision model using data from reported clinicaltrials.28 In their analysis, presented as a poster abstract, the cost-

Percent of Body Weight Lost at 12 Months in Each Treatment Group*

FIGURE 4

80.9

52.7

9.1

27.7

10.0

19.6

<5% 5%-9.9% ≥ 10%

90

80

70

60

50

40

30

20

10

0

n Nondrug

n Drug

% o

f P

opu

lati

on

* P < 0.001, χ2 = 44.5.


Weight-Loss Category



efficacy of orlistat was $531 per kilogram of weight loss versus$303 per kilogram of weight loss for sibutramine.

Recently, Warren and colleagues examined the incrementalcost utility of sibutramine compared with diet and lifestyleadvice for treatment of obesity.29 In this analysis, the cost-effective-ness of sibutramine was estimated using a decision model thataccounted for obesity-related cardiovascular disease and diabetes. The incremental cost per quality-adjusted life-year(QALY) for sibutramine was $9,299.

Two health technology assessment reports commissioned by theUnited Kingdom’s National Health Service evaluated sibutramineand orlistat independently.30,31 These studies reported the incremen-tal cost per QALY of sibutramine was £10,000 (approximately$18,700 in 2005) and £45,881 (approximately $85,750) for orlis-tat. There are no firm standards for an “acceptable” cost-effectivenessratio, but many analysts reference $50,000 per QALY or less as anacceptable threshold in the United States.32

Other studies have also found relationships between obesityand increased medical care costs.33-42 Most relevant to this study,Raebel et al. compared the participants in the LOSE Weightstudy with a matched control group of nonobese persons (BMI18.5 to 24.9).33 That analysis found that obese persons had agreater number of hospitalizations, prescription medications,professional claims, and outpatient visits than nonobese

persons. The median annual cost was significantly differentbetween the groups: $585 for obese persons and $333 fornonobese persons. Thompson et al. found that, in comparisonwith persons with a BMI of 20-24.9 kg/m2, costs for personswith a BMI of 25-29.9 kg/m2 and ≥ 30 kg/m2 were significantlyhigher for prescription drugs and for all types of medical care.34

Narbro et al. evaluated both the types and costs of medicationsmore often taken by obese individuals via a cross-sectional com-parison of the use of prescription medications in 1,286 obese indi-viduals in the Swedish Obese Subjects (SOS) study and 958 ref-erence individuals.39 Compared with nonobese persons, obese individuals had more medications for cardiovascular disease,pain, diabetes, and asthma. The overall cost of medications forobese individuals was more than 50% higher than for thenonobese population.

A retrospective database analysis of patients completing amembership health survey at Kaiser Permanente in northernCalifornia also demonstrated increased health care resource utilization and costs among obese members.40 Total health carecosts were increased in patients with a BMI of 30-34.9 kg/m2

and > 35 kg/m2. Mean annual costs were 25% greater amongthose with BMI of 30-34.9 kg/m2 and 44% higher among thosewith BMI ≥ 35 kg/m2 compared with those with BMI of 20-24.9 kg/m2 (P =0.003). More recently, Daviglus et al. found

Sibutramine Group (n=281) Nondrug Group (n=220) P Value*

Utilization Median (5th, 95th percentile) Median (5th, 95th percentile)

All physician visits 10 (3, 20) 8 (3, 20) 0.003

Obesity-related physician visits 5 (1, 9) 4 (1, 7) <0.001

All hospital visits 0 (0, 0) 0 (0, 0) 0.86

Obesity-related hospital visits 0 (0, 0) 0 (0, 0) 0.71

Professional service claims 0 (0, 16) 1 (0, 15) 0.57

Obesity-related professional service claims 0 (0, 1) 0 (0, 1) 0.56

Prescriptions obtained (new and refills) 10 (0, 38) 13 (1, 44) 0.05

Obesity-related prescriptions 0 (0, 14) 0 (0, 14) 0.32

Costs

Physician visit costs $326.98 (82.42, 1,115.30) $294.73 (69.87, 1,103.21) 0.14

Obesity-related physician costs $116.46 (23.29, 324.28) $82.42 (0, 252.40) <0.001

Hospital costs $0.00 (0.00, 0.00) $0.00 (0.00, 3,026.70) 0.80

Obesity-related hospital costs $0.00 (0.00, 0.00) $0.00 (0.00, 0.00) 0.87

Professional service costs $0.00 (0.00, 555.11) $5.25 (0.00, 801.09) 0.68

Obesity-related professional service costs $0.00 (0.00, 212.77) $0.00 (0.00, 72.22) 0.54

Prescription medication costs $384.74 (0.00, 2,468.86) $517.02 (2.47, 3,217.20) 0.02

Obesity-related prescription medication costs $0.00 (0.00, 873.48) $0.00 (0.00, 862.73) 0.33

Total health care expenditures $927.68 (136.17, 5,569.40) $1,031.98 (182.16, 6,617.95) 0.13

Obesity-related total health care expenditures $149.42 (23.29, 1,367.64) $126.76 (23.29, 1,470.99) 0.02

* All comparisons made using the Wilcoxon rank sum test.

Health Care Utilization and Costs for the 12-Month Period Before Study EnrollmentTABLE 2


that higher BMI values in young and middle-age persons wassignificantly associated with higher Medicare expenditures oncethese persons were older.41 Average annual charges for nonover-weight (BMI 18.5-24.9) women were $6,367 as compared with$11,985 for severely obese (BMI ≥ 35) women, with cumulativeexpenses being $76,866 and $174,752, respectively. Anotheranalysis of health care costs among the civilian noninstitution-alized population found that persons with a BMI between 18.5and 24.9 had mean annual expenditures of $2,970 comparedwith $4,333 for persons with a BMI ≥ 30.42

The decision to include antiobesity medications as a coveredbenefit by managed care organizations is a difficult one. Manyask to what extent the health plan is responsible for eating andother lifestyle habits of its members. However, it is clear that thelong-term consequences of obesity are substantial in terms of morbidity, mortality, and health care costs. Managed careorganizations must balance the up-front cost of technologieswith the likelihood that downstream costs will be borne byanother entity or the government. Currently, Medicare does notprovide reimbursement for obesity treatments, but a pressrelease from the Centers for Medicare and Medicaid Servicesindicates that obesity treatments may be covered in the future.43

Coverage for obesity treatments in the United States by managed care organizations is inconsistent, with few organizations

having clear clinical strategies for addressing weight management.44

A survey of HMOs found that 38.7% of plans covered prescriptionsfor anorexiants in 2003.45 In comparison, 50.9% of HMOs covered smoking cessation aids.

The study presented here evaluated sibutramine in the context of a structured WMP. The WMP portion of the studygave subjects the option of attending 2 of 20 American HeartAssociation seminars offered on weight management. In addition,participants were required to meet at least twice with a preventionspecialist. The structured WMP offered as a part of this study isconsistent with the notion that drug therapy alone is insufficientto produce significant weight loss. Payers willing to cover someor all of the cost of weight-loss drugs should do so with coveragecontingent upon participation in behavioral therapy in a WMP.

There are several strengths to the analysis presented here.First, unlike most economic analyses of obesity treatments, ourcost-effectiveness study was based solely upon clinical trial data.We did not model the lifetime cost-effectiveness. Other studiesextrapolate short-term weight loss into long-term changes inrisk for comorbid conditions associated with obesity. Thisapproach to extrapolation is tenuous because there has beenonly 1 study verifying these assumptions.39 Second, our studyincluded the full range of obese individuals, including subjectswho were older than 65 years (6%) and also subjects with BMIs

Health Care Utilization and Costs for the 12-Month Period After Study EnrollmentTABLE 3

Sibutramine Group (n=281) Nondrug Group (n=220) P Value*

Utilization Median (5th, 95th percentile) Median (5th, 95th percentile)

All physician visits 8 (2, 19) 7 (3, 19) 0.003

Obesity-related physician visits 4 (1, 8) 3 (0.5, 6) <0.001

All hospital visits 0 (0, 0) 0 (0, 0) 0.54

Obesity-related hospital visits 0 (0, 0) 0 (0, 0) N/A

Professional service claims 1 (0, 16) 1 (0, 15) 0.76

Obesity-related professional service claims 0 (0, 1) 0 (0, 1) 0.74

Prescriptions obtained (new and refills) 20 (4, 65) 17 (0.5, 58) 0.002

Obesity-related prescriptions 5 (1, 20) 0 (0, 15) <0.001

Costs

Physician visit costs $406.05 (90.82, 1,176.68) $353.22 (90.82, 1,205.93) 0.09

Obesity-related physician costs $158.93 (45.41, 412.16) $113.52 (22.70, 888.59) <0.001

Hospital costs $0.00 (0.00, 0.00) $0.00 (0.00, 0.00) 0.53

Obesity-related hospital costs $0.00 (0.00, 0.00) $0.00 (0.00, 0.00) N/A

Professional service costs $0.00 (0.00, 2,998.39) $0.00 (0.00, 2,939.43) 0.76

Obesity-related professional service costs $0.00 (0.00, 83.53) $0.00 (0.00, 61.93) 0.75

Prescription medication costs $435.12 (0, 4,250.61) $715.36 (2.43, 3,768.38) <0.001

Obesity-related prescription medication costs $0.00 (0.00, 873.48) $0.00 (0.00, 862.73) <0.001

Total health care expenditures $1,325.94 (131.85, 13,296.97) $1,485.83 (240.35, 11,601.46) <0.001

Obesity-related total health care expenditures $207.83 (45.41, 1,217.54) $191.05 (45.40, 1,619.89) <0.001

* All comparisons made using the Wilcoxon rank sum test.


> 40 (30%). Third, our study protocol did not require additional visits or services that were outside the scope of routine care for persons enrolled in a structured WMP. Theinterventions were delivered in the normal context of medicalcare delivery. Thus, our economic analysis did not have toadjust for protocol-induced utilization or atypical care. Finally,because the study was conducted within a single organization, itis likely that all health care utilization was captured, as com-pared with economic analyses that rely on patient self-report.

LimitationsSeveral limitations should be considered when interpreting theresults of our study. This analysis examined health care consumption over a 1-year period. Although the long-term consequences of obesity are costly in terms of medical resourcesconsumed, this study was of insufficient duration to fully capture these consequences. This limitation is evident in theabsence of measurable changes in both obesity-related and allhealth care resource utilization and costs between the preperi-od and intervention period (Table 4). Another significant limi-

tation of this study was the ability of randomization to ensurethat the groups were equal at baseline. This issue has been discussed in detail elsewhere.22 Briefly, the nonequivalence ofthe groups at baseline was related to the higher dropout rate of subjects randomized to receive only structured WMP. This study did not blind subjects to treatments, thus personswanting the medication to assist in weight loss were more likelyto disenroll from the study, creating nonequivalent groups. Thisdifference in baseline values was taken into account in the multivariate analysis, most notably because baseline BMI was anindependent predictor.

Another limitation of this study is that among patients randomized to sibutramine, the length of therapy was less than1 year, an average of 222 days per year (data not presented). Wedon’t know if the cost-effectiveness of therapy would improveor decline if the medication were used the full 365 days, but themajority of benefit obtained from the product appeared to berealized during the first 6 months of treatment.

When classifying medications and resource use as obesity-related or not, we did not attempt to reconcile differencesbetween the reviewers or arrive at a consensus. However, theintraclass correlation coefficients (a measure of level of agree-ment) ranged from 0.69 to 0.80, depending upon the type ofmedical service being evaluated. These values represent good tovery good agreement among the experts.

This study also examined only the direct costs of obesityfrom the perspective of a managed care organization. We didnot take into account indirect costs, such as absenteeism, lossof productivity, on-the-job injury, and premature mortality.Nonetheless, previous research has documented that, as BMIincreases, so do the number of sick days, short-term disability,and other indirect costs.46 Sturm asserts that, based on datafrom Health Care for Communities, a national household telephone survey in 1997-1998, the effects of obesity on physical health-related quality of life are similar to 30 years ofaging.47 Sturm further claims that obesity has a stronger associationwith reduced health-related quality of life and increased healthcare spending than does either problem drinking or smoking.47

Another limitation of this study was the use of national feeschedules for utilization of medical services and average whole-sale price for medications. Managed care organizations pay lessthan average wholesale price for medications, so this studyoverstated the actual managed care organization drug cost andprovides a conservative estimate of the cost-effectiveness of sibu-tramine. In addition, subjects receiving sibutramine were requiredto initially pay for the medication and then were reimbursed 75% of the cost to Kaiser Permanente. Previous research has shownthat reducing consumer cost sharing may improve weight loss.48

Thus, the true cost to the health plan was lower than the price usedin the calculations. Using contract prices or increasing the amountof consumer cost sharing for sibutramine is likely to improve thecost-effectiveness of this therapy.



Change in Health Care Resource Use and Expenditures for 12 Months Prior to Enrollment Compared With 12 MonthsAfter Enrollment

TABLE 4

Sibutramine Nondrug Group Group

Item (n=281) (n=220) P Value*

All Health Care Resource 5th, 95th 5th, 95th Use and Expenditures Median Percentile Median Percentile

Outpatient visits 1 0, 3 1 0, 3 0.80

Hospitalizations 0 0, 0 0 0, 0 0.45

Professional service claims 0 0, 3 0 0, 2 0.70

Prescription drugs 9 -5, 35 2 -9, 27.5 <0.001

Total costs ($) $1,279 $2,399, $271 -$4,217, <0.001$131,090 $63,840

Obesity-Related Health Care Resource Use 5th, 95th 5th, 95thand Expenditures Median Percentile Median Percentile

Obesity-related 1 0, 3 1 0, 2 0.67outpatient visits

Obesity-related 0 0, 0 0 0, 0 1.00hospitalizations

Obesity-related 0 0, 0 0 0, 0 0.38professional service claims

Obesity-related 3 0, 12 0 -2.50, 6 <0.001prescription drugs

Obesity-related costs ($) $408 -$6,077, $31 -$6,091, <0.001$4,868 $3,519

* Wilcoxon rank sum test.A negative number indicates less health care resource use in 12 months after

enrollment compared with 12 months prior to enrollment.



nn ConclusionThe incremental cost-effectiveness of sibutramine plus a structured WMP was $44 per additional pound of weight losscompared with a structured WMP alone. These results were primarily due to the cost of sibutramine and the additionalweight loss for those patients randomized to receive sibutramine. There was no observed cost savings in total healthcare resource use or expenditures in the sibutramine groupcompared with the nondrug group.

DISCLOSURES

Funding for this investigator-initiated study was provided by the KnollPharmaceutical Company, now owned by Abbott Laboratories, and was obtained by authors Marsha A. Raebel, Julie A. Porter, Frances A. Lanty,Elizabeth C. Gay, and John A. Merenich. The investigators collaborated withKnoll Pharmaceutical Company in protocol development and study initiation.Abbott Laboratories was given the right to review the manuscript, but theapproval of Abbott Laboratories was not required for publication of this manuscript. Merenich had participated in the speaker’s bureau for KnollPharmaceutical Company at the time of the study. Raebel, Porter, Lanty, Gay,and the other authors, Daniel C. Malone, Douglas A. Conner, and Erin A.Vogel, declare no financial interest in Abbott Laboratories. All authors discloseno potential bias or conflict of interest relating to this article.

All authors had access to study data, take responsibility for the accuracy ofthe analysis, and had authority over manuscript preparation and the decisionto submit the manuscript for publication. Malone served as principal authorof the study. Study concept and design were contributed by Porter, Raebel,Merenich, Lanty, Malone, Gay, and Vogel. Analysis and interpretation of datawere contributed by Malone, Raebel, Porter, Lanty, and Connor. Drafting ofthe manuscript was primarily the work of Malone and Raebel, and its criticalrevision was the work of all authors. Statistical expertise was contributed byMalone, Raebel, and Connor.

REFERENCES

1. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends inobesity among US Adults, 1999-2000. JAMA. 2002;288:1723-27.

2. Pi-Sunyer FX. Medical hazards of obesity. Ann Intern Med. 1993;119:655-60.

3. National Institutes of Health, National Heart, Lung, and Blood Institute.Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweightand Obesity in Adults: the Evidence Report. Obes Res. 1998;6(suppl 2):51S-209S.

4. Finkelstein EA, Feibelkorn IC, Wang G. National medical spending attributable to overweight and obesity: how much and who’s paying [Web exclusive]. Health Aff. 2003;W3:219-26.

5. Wolf AM, Colditz GA. Current estimates of the economic cost of obesity in the United States. Obes Res. 1998;6:97-106.

6. Wolf AM, Colditz GA. The cost of obesity: the US perspective.Pharmacoeconomics. 1994;5(suppl 1):34-37.

7. NHLBI Obesity Education Initiative Expert Panel on the Identification,Evaluation, and Treatment of Overweight and Obesity in Adults. ClinicalGuidelines on the Identification, Evaluation, and Treatment of Overweight andObesity in Adults: The Evidence Report. Bethesda, MD: National Institutes of Health; September 1998. NIH Publication 98-4083.

Cost-Effectiveness Analysis of SibutramineTABLE 5

Incremental Cost-Group Mean Cost ($) Mean Effect Cost/Effect* Effectiveness (95% CI)

Obesity-related costs/weight loss in poundsNondrug group (weight management only) 58 -5.1 pounds $12/pound lost $44/additional poundSibutramine group (sibutramine + weight management) 443 -13.7 pounds $32/pound lost lost ($44-$46)†

Obesity-related costs/percentage weight lossNondrug group (weight management only) 58 -2.2% $27/percent change

in body weightSibutramine group (sibutramine + weight management) 443 -6.0% $74/ percent change $101/additional

in body weight percent change in body weight($99-$102)†

Total health care costs/weight loss in poundsNondrug group (weight management only) $1,513 -5.1 pounds $297 /pound lostSibutramine group (sibutramine + weight management) $3,165 -13.7 pounds $230 / pound lost $194/additional

pound lost ($188-$200)†

* May not equal average cost/average effect due to rounding. † 95% confidence interval. CI = confidence interval.

Factors Predicting Total Direct Health Care Costs After Enrollment in LOSE Weight Study Subjects

TABLE 6

Beta Standard StandardizedItem Estimate Error Beta Estimate t Test P Value

Intercept -111.14 328.44 0.000 -0.34 0.74

Age 8.37 4.01 0.127 2.09 0.04

Gender 34.66 75.90 0.019 0.46 0.65

BMI -1.84 6.19 -0.017 -0.30 0.77

CDS -0.27 0.24 -0.419 -1.13 0.26

Study arm* 453.80 58.87 0.321 7.71 <0.001

BMI x CDS 0.01 0.004 0.224 1.73 0.08

Age x CDS 0.003 0.003 0.464 1.04 0.30

* Study arm equaled 0 for subjects receiving only weight management program and 1 for subjects receiving sibutramine plus weight management program.

BMI = body mass index; CDS =chronic disease score.



8. Brown CD, Higgins M, Donato KA, et al. Body mass index and the preva-lence of hypertension and dyslipidemia. Obes Res. 2000;8(9):605-19.

9. Stamler R, Stamler J, Riedlinger WF, Algera G, Roberts RH. Weight andblood pressure. Findings in hypertension screening of 1 million Americans.JAMA. 1978;240:1607-10.

10. Criqui MH, Mebane I, Wallace RB, Heiss G, Holdbrook MJ. Multivariatecorrelates of adult blood pressures in nine North American populations: theLipid Research Clinics Prevalence Study. Prev Med. 1982;11:391-402.

11. Dyer AR, Elliott P. The INTERSALT study: relations of body mass index to blood pressure. INTERSALT Co-operative Research Group. J Hum Hypertens.1989;3:299-308.

12. MacMahon SW, Blacket RB, Macdonald GJ, Hall W. Obesity, alcohol consumption and blood pressure in Australian men and women. The NationalHeart Foundation of Australia Risk Factor Prevalence Study. J Hypertens.1984;2:85-91.

13. Feinleib M. Epidemiology of obesity in relation to health hazards. AnnIntern Med. 1985;103:1019-24.

14. Garrison RJ, Castelli WP. Weight and 30-year mortality of men in theFramingham Study. Ann Intern Med. 1985;103:1006-09.

15. Gregg EW, Cheng YLJ, Cadwell BL, et al. Secular trends in cardiovasculardisease risk factors according to body mass index in U.S. adults. JAMA.2005;293(15):1868-74.

16. Park YW, Zhu S, Palaniappan L, Heshka S, Carnethon MR, Heymsfield SB.The metabolic syndrome: prevalence and associated risk factor findings in theU.S. population from the Third National Health and Nutrition ExaminationSurvey, 1988-1994. Arch Intern Med. 2003;163:427-36.

17. Colditz GA, Willett WC, Rotnitzky A, Manson JE. Weight gain as a risk fac-tor for clinical diabetes mellitus in women. Ann Intern Med. 1995;122:481-86.

18. Ford ES, Williamson DF, Liu S. Weight change and diabetes incidence: find-ings from a national cohort of U.S. adults. Am J Epidemiol. 1997;146:214-22.

19. Koh-Banerjee P, Wang Y, Hu FB, Spiegelman D, Willett WC, Rimm EB.Changes in body weight and body fat distribution as risk factors for clinicaldiabetes in U.S. men. Am J Epidemiol. 2004;159:1150-59.

20. Roth J, Qiang X, Marban SL, Redelt H, Lowell BC. The obesity pandemic:where have we been and where are we going? Obes Res. 2004;12(suppl 2):88S-101S.

21. Raebel MA, Conner DA, Porter JA, et al. The Long-term Outcomes ofSibutramine Effectiveness on Weight (LOSE Weight) study in a managed careorganization: twelve-month clinical outcomes in a naturalistic clinical setting[abstract]. American Diabetes Association 62nd Scientific Sessions; June 17,2002; San Francisco, CA. Diabetes. 2002;51(suppl 2):A413. Abstract 1695-P.

22. Porter JA, Raebel MA, Conner DA, et al. The Long-term Outcomes ofSibutramine Effectiveness on Weight (LOSE Weight) study: evaluating the roleof drug therapy within a weight management program in a group-modelhealth maintenance organization. Am J Manag Care. 2004;10(6):369-76.

23. Centers for Medicare and Medicaid Services. Available at: http://www.cms.hhs.gov/data/download/default.asp. Accessed June 5, 2003.

24. Bureau of Labor Statistics Consumer Price Indexes. Available at:http://www.bls.gov/cpi/. Accessed March 8, 2004.

25. Polsky D, Glick HA, Willke R, Schulman K. Confidence intervals for cost-effectiveness ratios: a comparison of four methods. Health Econ. 1997;6:243-52.

26. Clark DO, Von Korff M, Saunders K, Baluch WM, Simon GE. A chronicdisease score with empirically derived weights. Med Care. 1995;33(8):783-95.

27. Arterburn DE, Crane PK, Veenstra DL. The efficacy and safety of sibut-ramine for weight loss: a systematic review. Arch Intern Med. 2004;164:994-1003.

28. Bharmal M, Sankaranarayanan J, Viswanathan H, Shah B, Murawski M.Comparative cost efficacy analysis of sibutramine and orlistat in the treatmentof obesity: a third party payer’s perspective [abstract]. J Am Pharm Assoc.2002;42(2):312-13.

29. Warren E, Brennan A, Akehurst R. Cost-effectiveness of sibutramine in the treatment of obesity. Med Decis Making. 2004;24:9-19.

30. O’Meara S, Riemsma R, Shirran L, Mather L, ter Riet G. The clinical effec-tiveness and cost-effectiveness of sibutramine in the management of obesity: a technology assessment. Health Technol Assess. 2002;6:1-97.

31. O’Meara S, Riemsma R, Shirran L, Mather L, ter Riet G. A rapid and systematic review of the clinical effectiveness and cost-effectiveness of orlistatin the management of obesity. Health Technol Assess. 2001;5:1-81.

32. Siegel JE, Weinstein MC, Torrance GW. Reporting cost-effectiveness stud-ies and results. In: Gold MR, Siegel JE, Russell LB, Weinstein MC, eds. Cost-Effectiveness in Health and Medicine. New York, NY: Oxford University Press;1996:276-303.

33. Raebel MA, Malone DC, Conner DA, Xu S, Porter JA, Lanty FA. Healthservice use and health care costs of obese and non-obese individuals. ArchIntern Med. 2004;164:2135-40.

34. Thompson D, Brown JB, Nichols GA, Elmer PJ, Oster G. Body mass indexand future health care costs: a retrospective cohort study. Obes Res. 2001;9:210-18.

35. Allison DB, Zannolli R, Narayan KM, Venkat MD. The direct health carecosts of obesity in the United States. Am J Public Health. 1999;89:1194-99.

36. Oster G, Edelsberg J, O’Sullivan AK, Thompson D. The clinical and economic burden of obesity in a managed care setting. Am J Manag Care.2000;6:681-89.

37. Thompson D, Edelsberg J, Colditz GA, Bird AP, Oster G. Lifetime healthand economic consequences of obesity. Arch Intern Med. 1999;159:2177-83.

38. Oster G, Thompson D, Edelsberg J, Bird AP, Colditz GA. Lifetime healthand economic benefits of weight loss among obese persons. Am J PublicHealth. 1999;89:1536-42.

39. Narbro K, Ägren G, Jonsson E, Näslund I, Sjöstrom L, Peltonen M.Pharmaceutical costs in obese individuals: comparison with a randomly selectedpopulation sample and long-term changes after conventional and surgicaltreatment: the SOS intervention study. Arch Intern Med. 2002;162:2061-69.

40. Quesenberry CP, Caan B, Jacobson A. Obesity, health services use, andhealth care costs among members of a health maintenance organization. Arch Intern Med. 1998;158:466-72.

41. Daviglus ML, Liu K, Yan LL, et al. Relation of body mass index in youngadulthood and middle age to Medicare expenditures in older age. JAMA.2004;292:2743-49.

42. Wee CC, Phillips RS, Legedza AT, et al. Health care expenditures associatedwith overweight and obesity among U.S. adults: importance of age and race.Am J Public Health. 2005;95:159-65.

43. HHS announces revised Medicare obesity coverage policy: policy opens door to coverage based on evidence. Press release, July 15, 2004. Available at:http://www.hhs.gov/news/press/2004pres/20040715.html. Accessed March 15,2005.

44. Cook D. Coverage of obesity problematic for most health plans. ManagCare. 2002;11(7):41-44,46.

45. Managed Care Digest Series 2004. HMO-PPP/Medicare-Medicaid Digest.Bridgewater, NJ: Sanofi-Aventis; 2004.

46. Burton WN, Chen CY, Schultz AB, Edington DW. The economic costsassociated with body mass index in a workplace. J Occup Environ Med.1998;40:786-92.

47. Sturm R. The effects of obesity, smoking, and drinking on medical problemsand costs. Health Aff. 2002;21:245-53.

48. Hubbert KA, Bussey BF, Allison DB, Beasley TM, Henson CS, HeimburgerDC. Effects of outcome-driven insurance reimbursement on short-term weightcontrol. Int J Obes. 2003;27:1423-29.

cost-effectiveness of sibutramine in the lose weight study

Documents