Impact of 3-Tier Pharmacy Benefit Design and Increased Consumer Cost-sharing on Drug Utilization

Pamela B. Landsman, MPH, DrPH

Winnie Yu, PharmD

XiaoFeng Liu, PhD

Steven M. Teutsch, MD

Marc L. Berger, MD

The American Journal of Managed Care, October 2005, Volume 11, Issue 10

Objective: To estimate responsiveness of prescription demandwithin 9 therapeutic classes to increased cost-sharing comparedwith constant cost-sharing.

Study Design: Retrospective prescription claims analysis.

Methods: Between 1999 and 2001, 3 benefit plans changedfrom a 2-tier to a 3-tier design (cases); 1 plan kept a 2-tier design(controls). Study subjects needed 24 months of continuous coverageand a prescription filled ≥3 months before the benefit changefor a nonsteroidal anti-inflammatory agent (NSAID), a cyclooxygenase(COX-2) inhibitor, a selective serotonin reuptake inhibitor(SSRI), a tricyclic antidepressant (TCA), an angiotensin-convertingenzyme (ACE) inhibitor, a calcium-channel blocker (CCB), anangiotensin-receptor blocker (ARB), a statin, or a triptan. Changesin use were compared with the Wilcoxon signed rank test.Elasticity of demand among cases was calculated.

Results: Generally, medication possession ratios decreased forcases and increased for controls between 1999 and 2000. Switchrates increased for cases and decreased for controls for all classesbut CCBs. Switches to lower copayments for ACE inhibitors, statins,and triptans occurred more often for cases. Discontinuation-ratechanges for cases were 2 to 8 times those for controls. Generic-substitutionrates depended on availability and initial generic utilization.Elasticity of demand for drugs was generally low, -0.16 to-0.10, for asymptomatic conditions (ACE inhibitors, ARBs, CCBs,statins), and moderate, -0.60 to -0.24, for symptomatic conditions(COX-2 inhibitors, NSAIDs, triptans, SSRIs).

Conclusion: Use of retail prescription medications within 9specific therapeutic classes decreased as copayment increased.Demand for pharmaceuticals was relatively inelastic with thesecopayment increases.

(Am J Manag Care. 2005;11:621-628)

Because healthcare expenditures are expected tocontinue to rise, health insurers continue to seekways to slow the growth in spending by managingutilization and shifting costs to the consumers whilemaintaining a high level of customer satisfaction andquality outcomes. Although pharmaceutical spendingaccounts for only 7% of total healthcare expenditures,the rate of spending growth for prescription medicationshas outpaced other areas of medical care. In 2001prescription drug expenditures rose 16%, while those forhospital care and physician services rose 8% and 9%,respectively.1 Drug utilization review, prior authorization,generic substitution, and closed formularies aresome of the methods used by providers of pharmacybenefits to stem this tide. Most common among employersare incentive-based or multitiered formularies,where consumer cost-sharing increases for products onhigher tiers compared with lower tiers.

In 2000, 80% of health plans with prescription benefitsoffered 3-tier formularies compared with 36% ofplans 2 years earlier.2 In 2004 the average copaymentfor generic drugs (tier 1) was $10; preferred brandeddrugs (tier 2), $21; and nonpreferred branded drugs(tier 3) $33.1 Joyce et al estimated adding a third tier fornonpreferred brands to a 2-tier plan at a copayment of$30 decreases overall drug spending by 4% among planparticipants with employer-sponsored drug coverage.3Additional studies such as that by Huskamp et al haveshown changing from a 1-tier or 2-tier plan to a 3-tierplan impacts the purchasing decisions of consumersmore substantially with respect to their overall drugconsumption,4 but the impact of decreased utilizationon health outcomes has not been evaluated.4-6 Tworecent studies evaluated the impact of increasing prescriptiondrug copayments on uses of other healthcareservices. Motheral and Fairman found that enrolleeswho were moved from a 2-tier to a 3-tier pharmacy benefithad modestly lower prescription utilization with noincrease in physician office visits, emergency room visits,or hospital stays in the 11 months after the benefitchange.4 However, Goldman et al found that doublingthe copayments in a 2-tier plan for antidiabetic, antiasthmatic,and antiulcerant agents resulted in a 17%increase in predicted annual emergency departmentvisits and a 10% increase in predicted annual hospitaldays for persons with the respective conditions.7

Several studies from the 1980s and early 1990sshowed that demand for prescription drugs were highlyinelastic, with values between -0.33 and -0.10 for smallabsolute changes in price.8 As out-of-pocket costs andprice differentials between products on different tierscontinue to rise, consumers may be more sensitive tochanges in prescription drug costs than previouslyreported. This study uses a pre-post analysis of 3 managedcare populations whose pharmacy benefitschanged in mid-2000 from a 2-tier to a 3-tier design,compared with a managed care population that had nochange in a 2-tier benefit design during the same timeperiod. We evaluated changes in utilization patternssuch as medication possession ratio, discontinuationrates, and switches to lower-tier products, as well as theelasticity of demand for medications in 9 major therapeuticclasses.


Study Population

This study is a retrospective 2-sample cohort analysisof prescription medication utilization among managedcare enrollees. The study population (cases) wasmade up of 3 geographically dispersed managed careplans, each of which adopted a 3-tier pharmacy benefitsystem in the first half of 2000 (Table 1). Immediatelybefore the benefit switch, plans 1, 2, and 3 provided atraditional 2-tier prescription benefit covering genericand formulary brand products. Plans 1 and 3 mandatedthe new benefit for all members, whereas plan 2 adoptedit only for members of a preferred provider organization.Plan 4, the control population (controls), had thesame 2-tiered benefit structure as plan 2, but retainedthe 2-tier formulary during the evaluation period, 1999-2001. The analyses utilized enrollment data and pharmacyclaims for 1999 through 2001.

For inclusion in this analysis, members were firstrequired to have 24 months of continuous enrollment,12 months immediately before their plan benefit change(prechange period) and 12 months after their plan benefitchange (postchange period). If these members hada prescription filled during this time for medicationswithin any of the therapeutic areas of interest, theywere then assigned to 1 or both of 2 analysis groups.First, for each drug class of interest, members with aprescription filled at least 3 months before the end ofthe prechange period were retained for time-seriesanalyses related to drug utilization. Second, the subsetof cases with at least 2 prescriptions filled for a drugclass of interest within the 3 months before the benefitchange was retained for estimation of elasticity ofdemand.

Members were excluded from all analyses if theyobtained any medications through mail-order pharmacies(5%), as there were substantial differences betweencopayment rate and quantity of supply between mailorder and retail prescriptions. Additionally, 20% ofmembers whose copayment for a prescription wasinconsistent with their drug benefit were excluded fromanalysis. This group included those without a copaymentrecorded on their prescription claim or with acopayment amount recorded that did not match thecopayment tier as reported by the health plan. The latterwere excluded to attain a more accurate estimate ofthe impact of the copayment change on drug utilization.

Drug Classes

Several criteria were applied when selecting the drugclasses under study. The population of users of medicationswithin the drug class had to be of sufficient size tomake reasonable inferences about the population atlarge. The drug classes also had to be used to treat a varietyof chronic and acute, and symptomatic and asymptomaticconditions. Nine commonly used therapeuticclasses for 5 conditions wereselected for analysis:angiotensin-convertingenzyme (ACE) inhibitors, calcium-channel blockers (CCBs),and angiotensin-receptorblockers (ARBs) for cardiovascularconditions; cyclooxygenase2 (COX-2) inhibitors andnonselective nonsteroidal antiinflammatoryagents (NSAIDs)for pain; selective serotoninreuptake inhibitors (SSRIs)and tricyclic antidepressants(TCAs) for depression; 3-hydroxy-3-methylglutarylcoenzyme A reductaseinhibitors (statins) for lipid lowering; and serotonin 5-HT1 receptor agonists (triptans) for migraine.

Statistical Analysis

The prechange period refers to the 12 months precedingimplementation of the 3-tier benefit, and thepostchange period refers to the 12 months after implementationof the 3-tier benefit. Persistency outcomessuch as medication adherence (as measured by medicationpossession ratio), switching within a drug class, anddiscontinuation rates were measured for both theprechange and postchange periods among both cases andcontrols. Net changes in persistency measures for the 2populations then were compared by using the Wilcoxonsigned rank test. Two-sided tests with an alpha of .05were used for all study analyses to determine statisticalsignificance.

Binary indicators were created to allow tracking ofenrollees who were switched to another product withinthe same drug class during each 3-month interval for the4 quarters of the prechange and postchange periods.Overall switch rates and the proportion of these enrolleeswith a switch from a product with a higher copayment toone with a lower copayment are presented. Switchesfrom a brand name to a generic product also were examinedwhen a chemically equivalent generic was available.A drug was considered to be discontinued if a patient didnot fill a prescription for any drug in the same therapeuticclass for at least 30 days after completion of the lastprescription. Cumulative discontinuation rates for the6-and 12-month prechange and postchange periods forthe cases and the net changes at 12 months for both thecases and controls are presented.

Elasticity of Demand

To estimate elasticity of demand, the subset of caseswith at least 2 claims for any medication within theselected therapeutic areas during the 3 months beforethe benefit change was retained. Elasticity of demandwas defined as the ratio between the percent change inaverage monthly number of prescriptions filled and thepercent change in copayment from the prechange to thepostchange periods. Alternatively, elasticity of demandis the percent change in monthly prescription fills givena 1% increase in copayment. To adjust for differinglengths of time on therapy (because not all patientswere on therapy for the full 24 months under study),the average monthly number of prescriptions during theprechange and postchange periods—rather than thetotal number of prescriptions—was used to calculatepercent change in drug utilization.


Because drug classes consist of numerous productsthat usually require different copayments, the weightedaverage copayment for all products in a drug class representsthe average copayment rate for each class. Theprechange weighted average copayment was used toobtain the postchange copayment. The nominalcopayment is the amount that a patient would payunder the 3-tier system if he or she continued on thesame drug used in the prior period. Using the nominalcopayment was necessary as patients could switch tolower-copay products (within the same class) under the3-tier system; therefore, weighted average copayment inthe postchange period would not reflect the true magnitudeof copayment increase.


Population Characteristics

Information on enrollment and pharmacy benefitdesign for the 4 plans is presented in Table 1. Plans 1-3changed their pharmacy benefits from a 2-tier to a 3-tier structure with varying levels of copayment. Benefitsfor plan 4 and plan 2 were provided by the same insurer.Before the benefit change these 2 plans had the samebenefit design; however, after the benefit change plan 4retained its 2-tier formulary with the copayment levelsunchanged ($5 tier 1 and $10 tier 2). Plan 1 was comprisedof unionized employees of a single auto-industryemployer and their dependents. Plans 2, 3, and 4 werecomprised of employees of multiple employers in multipleindustries and their dependents.

Drug Persistence

Among case patients continuing on the same medicationor switching to another medication within thesame drug class after the benefit design change, statisticallysignificant but modest reductions in medicationpossession ratio (MPR) were observed (Figure 1). MPRchanges ranged from a 6.8% decrease in use of NSAIDsto a 1.7% increase in use of COX-2 inhibitors. For medicationsrelated to cardiovascular conditions (ACEinhibitors, ARBs, CCBs, statins), the MPR remainedover 80% at the end of the first year after the benefitchange. Conversely, MPR for those with no benefitchange increased for 8 of the 9 classes (range -0.1 to+6.8%). Statins showed a slight decrease (-0.1%) in MPRin the second year of the study period.


Cases had statistically significant changes in MPRcompared with controls in all but the ARB and COX-2classes. Differences in MPR for ARBs were greater forthe cases than for controls (-3.2% and +0.8%, respectively;> .05). MPR changes within the COX-2 classwere the only instance where MPR increased for thecases (+1.7%) and where the difference was greater forthe controls (+6.8%) between the prechange andpostchange periods. This difference, however, was notstatistically significant.

During the evaluation period, availability of genericmedications was limited to ACEs, CCBs, NSAIDs, andTCAs. Changes in generic substitution rate for ACEsand CCBs were greater among the cases than amongthe controls. Substitution rates increased among bothstudy populations for ACE inhibitors (5.5% cases and4.1% controls). Amongthose on CCBs, genericsubstitution decreasedfor controls (-3.7%)and increased for cases(+2.7%) in the yearsubsequent to the benefitchange. As mostcases and controls takingNSAIDs or TCAswere already on genericformulations beforeany benefit change,patients in bothgroups experiencedfew switches (near 0%)from branded to genericmedications.





Overall within-classswitch rates are presentedin Table 2. Casepatients switched productsan average of 1.5times more after thebenefit change comparedwith before thebenefit change and more than controls after the benefitchange for 8 of the 9 drug classes evaluated. Thisalso held true for the change in switch rates acrossthe 2 evaluation periods. However, only 3 of thesedifferences after the benefit change reached statisticalsignificance: statins (+3.8% cases vs -0.5% controls;= .001), NSAIDs (+4.8% cases vs -0.1%controls; = .011), and triptans (+7.5% cases vs -1.3%;= .024). Differences in switchrate for the cases compared withthe controls for the COX-2 classapproached statistical significance(+4.0% cases vs -3.2%; = .052).

Of those who switched productswithin a drug class, a significantlygreater proportion ofcases than controls switched to aproduct associated with a lowercopayment after the benefitchange for ACEs, statins, andtriptans (Figure 2). Switch-ratechanges to products with lowercopayments were nearly equalamong the study populations forARBs, CCBs, and SSRIs, whereasthe changes for NSAIDs andTCAs were greater among controls.These differences, however,did not reach statisticalsignificance.

Cumulative discontinuation rates for equivalent 6- and 12-month periods before and after the benefitchange for cases are presented in Table 3. For most ofthe drug classes, discontinuation rates increased substantiallyin the first 6 months after the copaymentchange compared with the same time frame in the previousyear. Over the 1-year period after the benefitchange, the majority of discontinuations for the casepopulation occurred in the first 6 months. The greatestchanges in discontinuation rates were among patientstaking ARBs, with a 25.7% net increase, and ACEinhibitors, with a 23.8% increase. The lowest netchanges were among users of COX-2 inhibitors (+4.1%)and triptans (-1.4%). Controls also had increased discontinuationrates after the benefitchange; however, all were less than10%. Changes in discontinuation ratesranged from 1.1% among controls takingstatins to 7.2% among controls takingARBs. The net changes indiscontinuation rates were greater forcases than controls in every drug classexcept triptans. For five of the classes—ACE inhibitors, ARBs, statins,SSRIs, and TCAs—the differencesbetween the groups were statisticallysignificant.

Elasticity of Demand

Elasticity of demand for each drugclass across the 3 plans is shown inTable 4 and Figure 3. Cases experiencedincreases in average copaymentsranging from 16% for TCAs to 129% forCOX-2 inhibitors. Overall, the highestelasticities were seen among drug classeswith the lowest prices before the benefitchange (TCAs and NSAIDs).Elasticity of demand for drugs treatingprimarily asymptomatic conditions (ACE inhibitors,ARBs, CCBs, statins) was relatively low, ranging from -0.16 to -0.10. Elasticities for ACEs (-0.14), ARBs (-0.16), and CCBs (-0.15) were similar in magnitude;statins had the lowest elasticity (-0.10). Responses toprice increases were greatest for prescription medications treating symptomatic conditions. The elasticity ofdemand for the 5 other drug classes evaluated rangedfrom -0.60 for NSAIDs to -0.24 for triptans. Price sensitivityfor NSAIDs was nearly twice that of COX-2inhibitors (-0.60 vs -0.31). Among antidepressants,the elasticity of demand for TCAs was the greatest at-1.15 and nearly 4 times greater than that for SSRIs(-0.27). Among these 3 plans, elasticities of demandwithin each drug class did not differ even after adjustingfor plan differences in age and sex (data not shown).


This is 1 of a small number of studies that haveexamined patterns of drug utilization and consumerprice sensitivity across a broad range of drug classesused to treat a variety of asymptomatic and symptomaticconditions. In this study we examined how drugutilization changed after prescription drug benefitplans changed from 2-tier to 3-tier designs. Utilizationpatterns were compared with those of an insuredcohort experiencing no change to their 2-tier benefitover the same time period. Overall, those experiencingan increase in tiers and copayments responded differentlythan those with no change in either tiers orcopayments. Increases in both groups appear to berelated to decreased medication possession ratios,increased switching to generics and other lower pricedalternatives within the same drug class, and increasedrates of discontinuation products within a drug class.

Consistent with findings in other published studies,the impact on medication adherence rates for patientswho elected to continue on their medication after abenefit change appeared to be small, with overall compliancerates of more than 80% for majority of therapeuticclasses examined.4-6 However, we also saw anegative impact on medication possession ratios forthose with a change in benefit compared with a smallincrease in possession ratios for those with no benefitchange. The increases in copayment also promoted agreater rate of switching to lower cost alternatives withina therapeutic class where available; the rate ofswitching depended on the type and number of alternativesavailable. Drug classes with generic or brandedalternatives with lower copayments that are equivalentor similar to the brand name product taken (eg, ACEinhibitors, CCBs) had higher rates of switching thanclasses without those therapeutic options (eg, ARBs,COX-2 inhibitors, statins).

Faced with an increase in copayment and tiers, upto one fourth of patients discontinued their medicationwithin the first 6 months after the benefit changerather than switch to another medication in the samedrug class, with the rates differing by therapeutic class.Those faced with no benefit change also had anincrease in discontinuation rates during the same timeperiod. However, there was no more than a 7.2%increase over the previous time period and little variationby therapeutic class. As adverse clinical consequencesof medication discontinuation may be severe,it is important to identify patients with true treatmentdiscontinuation as opposed to patients who switched toan alternative drug class to treat the same condition.

As summarized by the elasticity-of-demand estimatesfor each drug class, this study demonstrates thatpatients respond differently to an increase in their out-of-pocket costs for prescription medications dependingon the condition being treated, the absolute priceincrease, and the availability of treatment alternatives.Similar results were found by Goldman et al in ananalysis that simulated the doubling of copayments ina 2-tier benefit plan.7 Patients were most sensitive tocopayment increases for NSAIDs, which included COX-2 inhibitors. Our patient population was most sensitiveto copayment increases for NSAIDs specifically, followedby increases for COX-2 inhibitors. Price sensitivityfor NSAIDs was nearly twice that of COX-2inhibitors (-0.60 vs -0.31), due in part to the manylower priced prescription and over-the-counter optionsavailable within the NSAID class. Among persons takingantihypertensives, patients in both our study andthe one by Goldman et al were found to be among thoseleast responsive to rising copayments. In the study byGoldman et al, moderate sensitivity was estimated forpatients taking antihyperlipidemics (including statins)and lower sensitivity was estimated for patients takingantidepressants. Our patient population was moderatelysensitive to copayment increases for SSRIs to treatdepression and least sensitive to statins. The elasticityof demand for TCAs (utilized to treat depression) estimatedin our study is falsely high due to the relativelysmall increase in copayment for this drug class and thefact that the majority of patients were on generics priorto the benefit change. With the exception of the antidepressantresults, when Goldman et al modeled utilizationchanges among those with a diagnosiscorresponding to the treatment of interest, theyobserved patterns of sensitivity to copayment increasessimilar to the results presented here.

There are several limitations to this analysis, manyrelated to the use of administrative claims data forhealth services research. The copayment field for someclaims did not match the payment required by thepharmacy benefit plan, probably due to grandfatheringof benefit for some chronic users; deleting these claimsfrom the analysis has an unknown impact on theresults. The number of such claims was small; therefore,we believe the impact on our results should have beenminimal. Second, discontinuation rates may have beenoverestimated for NSAIDs and COX-2 inhibitors, as wellas ACE inhibitors and ARBs, because analyses wereconducted within a drug class and patients may havebeen switched to a product in an alternative but relatedprescription drug class for treatment of the same condition.In addition, for some classes (eg, triptans,NSAIDs), patients may have switched to an over-the-countertreatment or an alternative therapy. Any biases,however, should be similar for both study groups;therefore, the impact on our results should have beenminimal. The study also was limited solely to users ofretail pharmacies; enrollees using either mail-order andretail pharmacies or mail-order pharmacies exclusivelywere not examined. For elasticity of demand, enrolleeswere required to have 2 or more prescriptions withinthe selected therapeutic class before the benefit change.This was done to exclude enrollees who may have discontinuedbecause of problems with tolerance or whosecondition may have been resolved. Therefore, webelieve our elasticity estimates are conservative but stillhigh. Lastly, because patients are likely to utilize multipledrugs concurrently, elasticity-of-demand calculationsmay reflect increased sensitivity to the change intotal copayment burden rather than the price increaseof individual prescriptions.

It is reassuring that patients who choose to continuetheir medications appear adherent to them. Additionally,evaluation of the patterns of elasticity leads toconclusions similar to those of Goldman et al; patientsappear to make some rational decisions regarding medicationpurchases and trade-offs. The results showpatients are more price sensitive to medications used totreat primarily symptomatic conditions such as pain(chronic or acute), migraine, and asthma. Thesepatients have more alternative therapies available tothem, including over-the-counter options. Additionally,they may feel they can self-monitor their condition andbest decide when treatment is necessary. How successfulthese patients are needs to be evaluated.Alternatively, patients appear less sensitive to pricechanges among treatments for more "silent" conditionssuch as hypertension and hyperlipidemia.

This study has shown that demand for pharmaceuticalsin these populations was moderately inelastic andvaried by drug class. Elasticities for primarily chronic,asymptomatic treatments were lower than those for primarilyacute symptomatic treatments. Of greatest concernis the cohort of patients who choose to discontinuemedications following an increase in copayment. Ourfindings and those of other published studies report discontinuationrates as high as 25% following an increasein copayment regardless of the magnitude of theincrease. We further demonstrated a discontinuationrate of less than 10% when copayment remains constant.Therefore, copayment increases may lead to anadditional 15% of patients or more discontinuing theirmedication. Further research to examine patterns ofpatients switching among therapeutic classes is necessaryto fully capture the impact of increases in copayments.Additional studies are needed to determine whypatients make particular decisions regarding whichproducts to pay more for and which to discontinue.Additionally, the potential impact of discontinuation onhealth outcomes and healthcare utilization (eg, emergencydepartment visits, hospitalizations) needs to bequantified.

Modest increases in prescription copayments havebeen shown to have a negative impact on consumers'medication-purchasing decisions. These increases maylead to pill splitting or other reduced-dosing methods,increased time between refills, and increased medicationdiscontinuation, particularly for symptomatic medications,but also for classes of prescription medicationsused for long-term disease prevention. Shifting largerproportions of prescription costs to consumers may leadto unintended consequences such as increasing morbidity,mortality, and costs in other areas of the healthcaresystem.

From Wellpoint Pharmacy Management, West Hills, Calif (WY, XL); and OutcomesResearch & Management, US Medical & Scientific Affairs, Merck & Co, Inc, West Point, Pa(PBL, SMT, MLB).

This study was funded by Merck & Co, Inc.

Address correspondence to: Pamela B. Landsman, MPH, DrPH, Merck & Co, Inc, POBox 100, UG2B-74, 351 N Sumneytown Pike, North Wales, PA 19454-2505.

1. The Kaiser Family Foundation. Prescription drug trends. Fact sheet #3057-03.Available at: Accessed September 13,2005.


Health Benefits 2003 Annual Survey.

2. Kaiser Family Foundation and Health Research Educational Trust. Kaiser Family Foundation and HealthResearch Educational Trust, ed. Menlo Park, Calif: 2003.


3. Joyce GF, Escarce JJ, Solomon MD, et al. Employer drug benefit plans andspending on prescription drugs. 2003;288:1733-1739.

N Engl J Med.

4. Huskamp HA, Deverka PA, Epstein AM, et al. The effect of incentive-based formularieson prescription-drug utilization and spending. 2003;349:2224-2232.

Med Care.

5. Motheral BR, Fairman KA. Effect of a three-tier prescription copay on pharmaceuticaland other medical utilization. 2001;39:1293-1304.

Med Care.

6. Rector TS, Finch MD, Danzon PM, et al. Effect of tiered prescription copaymentson the use of preferred brand medications. 2003;41:398-406.


7. Goldman DP, Joyce GF, Escarce JJ, et al. Pharmacy benefits and the use ofdrugs by the chronically ill. 2004;291:2344-2350.


8. Smith DG. The effects of copayments and generic substitution on the use andcosts of prescription drugs. 1993;30:189-198.