Budget Impact of Tegaserod on a Managed Care Organization Formulary

April 15, 2005
Michael A. Bloom, BA

,
Victoria Barghout, MSPH

,
Kristijan H. Kahler, SM

,
Judith Bentkover, PhD

,
Hannah Kurth, BA

,
Ian M. Gralnek, MD, MSHS

,
Brennan M. R. Spiegel, MD, MSHS

Supplements and Featured Publications, Economic Impact of Irritable Bowel Syndrome: What Does the Future Hold?, Volume 11, Issue 1 Suppl

We sought to develop a budget impact modelthat assesses the economic effect of addingtegaserod for the management of irritable bowel syndrome(IBS) with constipation to the formulary of amanaged care organization (MCO). The model estimatesthe per patient economic impact and the permember, per month (PMPM) economic impact ofpatients 6 months before and 6 months after the initiationof tegaserod. Resource utilization data, takenfrom medical and pharmacy administrative claimsdata, were based on a retrospective, longitudinalstudy of 3365 patients administered tegaserodthrough a large, geographically diverse MCO. Costswere estimated for 2 patient subgroups, women withIBS and other gastrointestinal (GI) diagnoses.Sensitivity analyses were performed by varying severalmodel input parameters. The base-case modelresulted in an incremental PMPM budget impactassociated with the use of tegaserod of $0.01. Totalper patient budget impact (for all resources, includingtegaserod) for a 6-month period was $274.34 forwomen with IBS and $301.84 for women with otherGI diagnoses. Overall, 25.9% (29.0% for womenwith IBS group and 21.9% for women with other GIdiagnoses group) of the cost of tegaserod was offsetby decreases in resource utilization. Key drivers ofpost-tegaserod reductions in resource costs werehospital stays, outpatient office visits, emergencydepartment visits, endoscopic procedures, andnonendoscopic procedures. Tegaserod therapy candecrease GI-related resource utilization, resulting ina significant cost-offset percentage. When the associatedbudget impact of adding tegaserod to its formularyis absorbed across an entire MCO population,the PMPM impact of tegaserod is small.

(Am J Manag Care. 2005;11:S27-

S34)

Irritable bowel syndrome (IBS) is characterizedby symptoms of abdominal painor discomfort associated with altered/disturbedbowel movements.1 IBS is highlyprevalent, with estimates in the US generalpopulation ranging from 10% to 20%.2 Mostpeople with IBS are women.3

Despite the large number of patients withIBS and the unpleasantness of its symptoms,only a small number of patients with IBSseek medical attention.4 Although the numberof patients seeking treatment is small,IBS results in a considerable number ofphysician visits and pharmacologic treatments.Approximately 12% of patients whoconsult primary care physicians and 28% ofpatients who consult gastroenterologists areseeking help for symptoms of IBS.5

Aside from the physical and emotionaltoll, IBS results in a significant economicburden to society, with annual costs in theUnited States estimated by various studiesto be $1.7 billion to more than $10.5 billionin direct medical costs (excluding prescriptionand over-the-counter drug costs) and$20 billion for indirect costs,6 which resultfrom decreased productivity at work andschool.7,8 The economic impact of IBS issimilar to or higher than that of other long-termconditions, such as hypertension, asthma,Crohn's disease, chronic liver disease,and cirrhosis.9,10

Health status and health-related quality oflife (QOL) of patients with IBS are poorcompared with those of the general population.11-15 Furthermore, several studies haveshown that the health-related QOL ofpatients with IBS is similar to or lower thanthat of patients with other debilitating conditions,such as gastroesophageal reflux disease,asthma, migraine, diabetes mellitus,and end-stage renal disease.14,16

Although most patients with IBS withconstipation (IBS-C) are initially treatedwith fiber supplements and other nonpharmacologictherapies,17 many do not achieveadequate clinical response to these treatments.3 Alternatives to fiber supplementsare now available for the treatment of IBS. Arelatively newer agent to join the armamentariumof therapies to treat gastrointestinal(GI) motility disorders is tegaserod maleate,a 5-hydroxytryptamine type 4 (5-HT4) receptoragonist. Tegaserod maleate is the only USFood and Drug Administration-approvedagent for the treatment of women with IBS-Cand for the treatment of patients (men andwomen) younger than 65 years of age whohave chronic idiopathic constipation. Tegaseroddemonstrates clinical efficacy in reducingsymptoms associated with IBS-C,including abdominal pain, bloating, andinfrequent bowel movements. Tegaserod hasa favorable safety and tolerability profile andhas demonstrated efficacy in providing globalsymptom relief.18,19

Demonstration of clinical efficacy inpatients with IBS is arguably the mostimportant criterion used by physicians andpayers, such as managed care organizations(MCOs), when selecting a particular therapy.Intense scrutiny has recently beenplaced on MCOs with regard to coverage forselect therapeutic areas. In part because ofthis new development, several studies havebeen conducted to demonstrate the overalleconomic impact associated with IBS.20,21 Todate, however, no study has considered thepotential budgetary impact of introducingtegaserod therapy to an MCO formulary.Therefore, we sought to develop a budgetimpact model in an MCO to compare GI-relatedresource utilization and costs 6months before and after the initiation oftegaserod therapy for tegaserod users andnonusers. We hypothesized that theincreased costs of tegaserod compared withalternative therapies would be partially offsetby downstream savings related to a decreasein GI-related resource utilization engenderedby the effectiveness of tegaserod.

Methods

Model Structure.

The budget impactmodel was developed to simulate the 6-month budgetary impact of adding tegaserodto an MCO formulary. The model estimatesresults for 2 patient subgroups based onpatient diagnosis: women with IBS and otherGI diagnoses, which include abdominal pain,intestinal disorders, GI/digestive disorders,liver/pancreatic disorders, GI surgical disorders,and GI cancer.

This budget impact model takes the simpleapproach of estimating the costs of thewomen with IBS and other GI diagnosesgroup before the introduction of tegaserodand the costs of women with IBS andother GI diagnoses after the introduction oftegaserod. Costs before the introductionof tegaserod were considered the burden ofillness of the group of women with IBS andthe other GI diagnoses group and were estimatedby combining the relevant resourceutilization with the costs of those resources.To determine the costs of womenwith IBS and GI diagnoses after the introductionof tegaserod, the burden of illness ofwomen with IBS and GI diagnoses wasadjusted by the relative impact of tegaserodon those resources.

Resource Utilization Study.

Data to populatethe model came from a retrospective,longitudinal resource use study of patientsadministered tegaserod from a large, geographicallydiverse MCO covering 14 millionlives; medical and pharmacy administrativeclaims data were used to identify suchpatients from August 1, 2002 (launch oftegaserod), through June 30, 2003.22 Theresource utilization study population consistedof 3365 patients taking tegaserod andcompared the relative GI resource utilizationin the 6 months after tegaserod initiationwith that in the 6 months beforetegaserod initiation.

Patient Population.

In the budget impactmodel, we used a default patient populationof 10 million as determined by data from theMCO that conducted the resource utilizationstudy.

Prevalence of IBS in Women and ofOther GI Diagnoses

We defined the prevalence of IBS inwomen as the proportion of covered lives ofwomen who had a diagnosis of IBS. Wedefined the prevalence of the other GI diagnoses subgroup as the proportion of coveredlives of patients with other GI diagnoses,excluding those in the women with IBS subgroup.Prevalence rates from the aforementionedresource utilization study were 1.2%for the women with IBS subgroup and 26.7%for the other GI diagnoses subgroup, basedon the observed MCO prevalence rates,22which became the default values in thebudget impact model.

For both patient subgroups, we definedthe tegaserod treatment rate as the proportionof women with IBS or other GI diagnosespatients who received at least 1prescription for tegaserod. Again, based onobserved MCO treatment rates,22 we used anestimated tegaserod treatment rate of 2% forthe women with IBS subgroup and 0.1% forthe other GI diagnoses subgroup in themodel.

We selected pharmacy and nonpharmacyresource utilization categories based on aprevious classification scheme used byLongstreth and colleagues.23 Categories andindividual items were specified in an a priorimanner without knowledge of the actualinformation in the medical and pharmacydatabase. The final medical resource categorizationselection was reviewed, revised, andapproved by an advisory team of specialistswith experience in gastroenterology, functionalGI disorders, and health servicesresearch.

Information on the relative impact oftegaserod on GI-related resources wasobtained from the resource utilization studythat assessed the following outcomes: physicianoffice visits, hospitalizations, emergencydepartment visits, endoscopic and nonendoscopicprocedures and GI medications (IBSdrugs, inflammatory bowel disease drugs,proton pump inhibitors [PPIs], promotilityagents, ulcer drugs, antispasmodics, H2 antagonists,antidiarrheals, and laxatives). Anymedical resource utilization category thatshowed use of less than 5 per 1000 patientswas considered too unstable for inclusion inthe model and was set to zero.

Costs.

GI-related costs for the categoriesof service listed above were determined byusing the resource-based relative value scalefor professional services and gap codes (partof the resource-based relative value scale) forservices not requiring physician intervention.Inpatient costs were determined usingthe 2003 payment reference average nationalpayments rates for diagnosis-relatedgroups, and drug costs used a 20% discount ofthe average wholesale price for costs.24

The cost of tegaserod was based on actualutilization in the MCO. A mean of 2.27 prescriptionswas filled by patients per 6months of follow-up, with a mean of 68 days'supply.22

Budget Impact Model Analyses.

Themodel was developed from the perspective ofan MCO. The time horizon considered forthis analysis was 6 months; therefore, dollarvalues were not discounted. For each patientsubgroup (women with IBS and other GIdiagnoses), the 6-month per patient budgetimpact attributable to tegaserod was calculatedby subtracting the post-tegaserodintroduction pharmaceutical and medicalGI-related costs from pre-tegaserod introductionper patient costs.

We calculated the per member, permonth (PMPM) budget impact by dividingthe 6-month per patient budget impact by 6,multiplying the resultant PMPM budgetimpact by the number of patients, and dividingby the total number of members.

We calculated a cost-offset percentage forboth patient subgroups by dividing the total6-month impact in non-tegaserod resourceutilization costs by the 6-month per patientcost of tegaserod. The cost-offset percentagerepresented the percentage of tegaserodcosts that can be offset by reductions inother resource utilization.

Sensitivity Analyses.

We performed sensitivityanalyses varying the inputs in themodel to observe their effects on the totalcost PMPM and the tegaserod cost-offsetpercentage.

Results

Budget Impact Analysis.

The pretegaserodintroduction of nonpharmaceutical,GI-related resource utilization defaultvalues are presented in Table 1.

In the budget impact model, the default 6-month cost of tegaserod was $386.44 perpatient (Table 2). The total per patientbudget impact (for all resources, includingtegaserod) for a 6-month period was $274.34for women with IBS and $301.84 for otherGI diagnoses. This translated to a PMPMbudget impact of $0.01 for each patientgroup (Table 3).

Key drivers of post-tegaserod reductionsin resource costs for women with IBS werehospital stays (-$38.16), abdominal andpelvic computed tomography (-$28.03),colonoscopy (-$27.78), and outpatientoffice consultations (-$18.20). Women withIBS realized small reductions in costs associatedwith emergency room visits (-$9.65),upper GI endoscopy (-$6.63), antispasmodics(-$4.78), and H2 antagonists (-$2.64).Key drivers of post-tegaserod reductions inresource costs for other GI diagnoses wereslightly different, with key drivers ofresource costs post-tegaserod consisting ofhospital stays (-$62.32), emergency departmentvisits (-$30.76), outpatient office consultations(-$28.43), and abdominal andpelvic computed tomography ($22.74).Patients with other GI diagnoses had smallerreductions in costs associated withabdomen ultrasound (-$9.79) and upper GIendoscopy.

Overall, 25.9% (29.0% for women with IBSand 20.9% for other GI diagnoses) of the costof tegaserod was offset by associateddecreases in other resource utilization for alltegaserod patients (Figure 1).

Sensitivity Analyses.

We varied theprevalence of women with IBS and other GIdiagnoses to determine the effect on the 6-month tegaserod cost per patient. For eachpercentage increase in the prevalence rate,an equivalent PMPM percentage increase integaserod cost was observed. For example,when the prevalence rates for women withIBS and other GI diagnoses were increased50% from 1.2% to 1.7% for women with IBSand from 26.7% to 40% for other GI diagnoses,the total cost PMPM increased from $0.011 to$0.016 for women with IBS and from $0.009to $0.014 for other GI diagnoses.

Similar results were seen when varyingthe prevalence of women with IBS and otherGI diagnoses to determine the effect on the6-month tegaserod cost per patient. Foreach percentage increase in the prevalencerate, an equivalent percentage increase integaserod cost per patient was seen. Whenthe tegaserod treatment rate for women withIBS was increased 50% to 3.1%, the total costPMPM increased from $0.011 to $0.016.When the tegaserod treatment rate for otherGI diagnoses was increased 50% from 0.067%to 0.1%, the total cost PMPM increased from$0.009 to $0.014.

The 6-month tegaserod cost per patientwas varied to examine the effect on cost offset,and the relationship between the 2 isshown in Figure 2 for women with IBS andin Figure 3 for other GI diagnoses. As thetegaserod cost per patient decreases, thecost offset increases by the same percentage.In other words, if the cost of tegaserod isreduced to half ($193.22), the cost offsetdoubles to 58% for women with IBS and43.8% for other GI diagnoses. For the cost oftegaserod to be completely (100%) offset byreductions in other resources, the tegaserodcost per patient must be decreased to$112.09 for women with IBS and $84.56 forother GI diagnoses. Varying the percentagesof patients with women with IBS and otherGI diagnoses or the tegaserod treatment ratedid not affect the cost-offset percentagebecause all end points examined were perpatient amounts.

Discussion

Our analysis indicates that introducingtegaserod to an MCO formulary candecrease GI-related medical costs in womenwith IBS and other GI diagnoses comparedwith costs before the approval of tegaserod.When these GI-related resources and theirassociated costs are considered in additionto the cost of the drug, the impact of introducingtegaserod to an MCO formulary issmall ($0.01 PMPM). Nearly one third (29%)of the cost of tegaserod may be offset byreductions in costs in other resource usecategories for women with IBS. For other GIdiagnoses, more than one fifth (21.9%) of thecost of tegaserod is offset by reductions inassociated resources. The results of themodel were robust, as evidenced by our discoverythat varying the percentage ofwomen with IBS and with other GI diagnosesor varying the tegaserod treatmentrate did not affect the cost-offset percentage.

The cost of tegaserod is only partially offsetby reductions in medical resources; however,because of its ability to providemultiple symptom relief, tegaserod may be abetter treatment than other single-symptomoptions, such as laxatives and antispasmodics.Because other therapies do nottarget all of the symptoms associated withIBS (ie, abdominal pain, bloating, and constipation),patients may attempt to obtainrelief for all of their symptoms by using multipletreatments.

This model demonstrated that introducingtegaserod in women with IBS andpatients with other GI diagnoses is associatedwith a PMPM increase of $0.01. Thisnumber is very small given the averageMCO PMPM expenditures for other commonIBS treatments, such as PPIs ($2.55) andH2 antagonists ($0.23), and for medications,such as antidepressants ($3.22), antidiabeticagents ($1.33), contraceptive hormones($0.74), and antimigraine agents ($0.44),prescribed for other prevalent conditions.25

Several studies of IBS patients in theUnited States, including those in the managedcare setting, demonstrate that IBSresults in a significant economic impact.9,10,21,26,27 This current model demonstratesthat treatment of IBS with tegaserodwill help to offset some of these increasedcosts, using a budget impact model that isdirectly relevant to payers who may potentiallybe including this product in their formularies.

In addition to the impact on direct medicalcosts, tegaserod may result in other benefitsnot examined in this study, such asimproved health-related QOL and increasedproductivity at work, school, and home. Inparticular, one study shows that indirectcosts associated with lost wages anddecreased productivity account for thelargest proportion of the total IBS burden7;other studies show a substantial IBS burdenin the form of indirect costs, such as medicallyrelated absenteeism.6,26,27 The resultsof these studies support the need for futurestudies to assess the indirect cost benefitassociated with tegaserod.

This model demonstrated that PMPMcosts might increase slightly when tegaserodis introduced in women with IBS, the currentindication for tegaserod. However, similarto the results of the group of women withIBS, the budget impact model also demonstrateda low PMPM cost when tegaserod wasintroduced in the other GI diagnoses group.Therefore, it is anticipated that futureresearch may demonstrate the usefulnessand small budget impact of tegaserod in theother GI diagnoses group, such as chronicconstipation.

Although this model can be used as a toolto assess the economic impact of IBS, it is notwithout its limitations. First, as in any model,the accuracy of the data input into the modeldictates the accuracy and credibility of thefinal end product of the model.28,29 Therefore,limitations that may affect the data from theresource use study22 include limited studyduration and sample size, selection bias relativeto patients chosen to be treated withtegaserod compared with other therapies, andlack of control for sampling/drug supply comparedwith actual filled prescriptions.Collectively, this may affect the generalizabilityof results. Another potential drawback ofhow the model is implemented is the use ofnational cost estimates. A model that was customizedto individual MCO plans might provemore beneficial and might provide a moreaccurate cost assessment for individual plans.

Conclusion

This model provides evidence that introducingtegaserod therapy in a managed carepopulation may result in decreases in GI-relatedresource utilization costs and in relativelysmall increases in overall PMPMcosts. The results of this study may be usefulto any MCO considering adding tegaserodto its formulary.

In addition, this model demonstrates thata portion (25.9%) of the cost of treating withtegaserod could be offset by associatedreductions in other resource categories(pharmacy, inpatient, outpatient, endoscopic,and nonendoscopic resources). When theassociated budget impact is absorbed acrossan entire MCO population, the PMPM impactof tegaserod is very small.

Gastroenterology.

1. Drossman DA, Camilleri M, Mayer EA, WhiteheadWE. AGA technical review on irritable bowel syndrome.2002;123:2108-2131.

Curr Gastroenterol Rep.

2. Cash BD, Chey WD. Advances in the management ofirritable bowel syndrome. 2003;5:468-475.

Am J Gastroenterol.

3. American College of Gastroenterology FunctionalGastrointestinal Disorders Task Force. Evidence-basedposition statement on the management of irritable bowelsyndrome in North America. 2002;97(suppl):S1-S5.

Prim Care.

4. Browning SM. Constipation, diarrhea, and irritablebowel syndrome. 1999;26:113-139.

Medscape Gen Med

5. Ringel Y, Drossman DA. Toward a positive and comprehensivediagnosis of irritable bowel syndrome.[serial online]. 2000;4(2). Availablefrom: WebMD Medscape Health Network, New York,NY. Accessed November 18, 2004.

J

Manag Care Pharmacy.

6. Hulisz D. The burden of illness of irritable bowel syndrome:current challenges and hope for the future. 2004;10:299-309.

Am J Manag Care.

7. Martin R, Barron JJ, Zacker C. Irritable bowel syndrome:toward a cost-effective management approach.2001;7(suppl):S268-S275.

Dig Dis

Sci.

8. Drossman DA, Li Z, Andruzzi E, et al. US householdersurvey of functional gastrointestinal disorders: prevalence,sociodemography, and health impact. 1993;38:1569-1580.

Curr Med

Res Opin.

9. Martin BC, Ganguly R, Pannicker S, Frech F,Barghout V. Utilization patterns and net direct medicalcost to Medicaid of irritable bowel syndrome. 2003;19:771-780.

Gastroenterology.

10. Sandler RS, Everhart JE, Donowitz M, et al. Theburden of selected digestive diseases in the UnitedStates. 2002;122:1500-1511.

World J Gastroenterol.

11. Si JM, Yu YC, Fan YJ, Chen SJ. Intestinal microecologyand quality of life in irritable bowel syndromepatients. 2004;10:1802-1805.

Pharmacoeconomics.

12. Akehurst RL, Brazier JE, Mathers N, et al. Healthrelatedquality of life and cost impact of irritable bowelsyndrome in a UK primary care setting.2002;20:455-462.

Ann Intern Med.

13. Creed F, Ratcliffe J, Fernandez L, et al. Health-relatedquality of life and health care costs in severe, refractoryirritable bowel syndrome. 2001;134:860-868.

Gastroenterology.

14. Gralnek IM, Hays RD, Kilbourne A, Naliboff B,Mayer EA. The impact of irritable bowel syndrome onhealth-related quality of life. 2000;119:654-660.

Digestion.

15. Hahn BA, Yan S, Strassels S. Impact of irritablebowel syndrome on quality of life and resource use inthe United States and United Kingdom. 1999;60:77-81.

Clin Ther.

16. Frank L, Kleinman L, Rentz A, Ciesla G, Kim JJ,Zacker C. Health-related quality of life associated withirritable bowel syndrome: comparison with other chronicdiseases. 2002;24:675-689.

N Engl J Med.

17. Mertz HR. Irritable bowel syndrome. 2003;349:2136-2146.

Scand J Gastroenterol.

18. Nyhlin H, Bang C, Elsborg L, et al. A double-blind,placebo-controlled, randomized study to evaluate theefficacy, safety and tolerability of tegaserod in patientswith irritable bowel syndrome. 2004;39:119-126.

Aliment Pharmacol Ther.

19. Novick J, Miner P, Krause R, et al. A randomized,double-blind, placebo-controlled trial of tegaserod infemale patients suffering from irritable bowel syndromewith constipation. 2002;16:1877-1888.

Gastroenterology.

20. Camilleri M. Management of the irritable bowel syndrome.2001;120:652-668.

J Clin Gastroenterol.

21. Patel RP, Petitta A, Fogel R, Peterson E, ZarowitzBJ. The economic impact of irritable bowel syndrome ina managed care setting. 2002;35:14-20.

Am J Manag Care.

22. Stephenson JJ, Barghout V, Kahler KH, et al. Theeffectiveness of tegaserod therapy on GI-related resourceuse in a managed care population. 2005;(suppl)11:S35-S42.

Am J Gastroenterol.

23. Longstreth GF, Wilson A, Knight K, et al. Irritablebowel syndrome, health care use, and costs: a US managedcare perspective. 2003; 98:600-607.

Red Book: 2004 Drug Topics.

24. Montvale, NJ: MedicalEconomics Company; 2004.

Pharmacy Benefit Report Facts and Figures.

25. EastHanover, NJ: Novartis Pharmaceuticals; 2003.

Aliment

Pharmacol Ther.

26. Inadomi JM, Fennerty MB, Bjorkman D. The economicimpact of irritable bowel syndrome. 2003;18:671-682.

Arch Intern Med.

27. Leong SA, Barghout V, Birnbaum HG, et al. Theeconomic consequences of irritable bowel syndrome: aUS employer perspective. 2003;163:929-935.

Gastroenterology.

28. Spiegel BM, Targownik LE, Kanwal F, et al. Thequality of published health economic analyses in digestivediseases: a systematic review and quantitativeappraisal. 2004;127:403-411.

Arch Intern Med.

29. Spiegel BMR, Gralnek IM, Bolus R, et al. Clinicaldeterminants of health-related quality of life in patientswith irritable bowel syndrome. 2004;164:1773-1780.