Effect on Drug Utilization and Expenditures of a Cost-Share Change From Copayment to Coinsurance

Abstract

BACKGROUND: Whileincreasesinprescriptiondrugspending have moderated in recentyears,drugspending is still aconcern among managed careorganizations and healthplanadministrators .Inorder to minimize costincreasesfromyeartoyear, many healthcare plans have shiftedmoreofthe costofmedications to themember-consumer. Coinsurance,abenefitdesign in which thepatient pays apercentage of thecostofthe medication, is garnering more attentionasatypeof cost-sharing that differsfromthe traditionalcopaymentmodel.OBJECTIVE: To estimate theimpactonmedicationexpenditures and utilization of apharmacybenefit design changefrom3-tiercopaymentto coinsurance.METHODS:Drugexpenditures and utilization of beneficiaries aged ≥ 18 yearsand continuously enrolled in 2privately insured groupswere compared beforeand afterabenefit design changein1ofthe groups. For the12monthsbeforethe benefitdesign change, both groupshad a3-tier, fixed-dollarcopaymentstructurewithidentical cost-sharing per 30-day supply:$10 tier-1 copaymentfor genericdrugs,$25 tier-2 copaymentfor preferred branddrugs,and $40tier-3copaymentfor non-preferred brand drugs. On September 1, 2005,a4-tiercoinsurance benefitdesign (25% forall tiersexcepttier-3[non-preferred]drugs at 50%, with minimum and maximum patient out-of-pocket[OOP]costapplied to each tier)was implemented in theinterventiongroup (N=46,311). The3-tiercopayment design wasmaintainedinthe comparison group (N=7,916). Adifferencein-difference analysis wasusedtoestimate theeffectofthe benefitdesign changeonexpenditures and utilization,overall (for allprescriptiondrugs), and for3classes of essentialmedications:antihypertensives,antidepressants, andstatins.Analyses measured changes in outcomesfrom6months pre-change (October 1, 2004,through March31, 2005)through 6months post-change (October1,2005, throughMarch 31,2006. In theoverall (all drug)analyses,per memberper month (PMPM) outcome measures were totalpharmacyclaims and cost, beneficiary (patientOOP)cost, and employer(plan sponsor)cost. Analyses of the3essential drug classes were limited to members with at least 1claiminthe drug class in both the pre-changeand post-changeperiods (N=11,917, intervention group;1,792, comparison group), and assessedper patient per month (PPPM)dayssupply,beneficiary cost, employercost, and total cost. RESULTS:Beneficiaries in theinterventiongroup paid31.8% of total pharmacybenefit cost at thepoint of care versus 31.5%inthe comparison group in thepost-changeperiod. Theincreasesinbeneficiary costfromthe pre-changeperiodtothe post-changeperiodwerenot significantlydifferentfor theintervention(7.5%)and comparison (3.0%) groups(P =0.983). From thepre-changeperiodtothe post-changeperiod, total spending per memberincreased $4.57PMPM(6.3%), from$72.29to$76.87inthe intervention group versus a$5.87 PMPM increase (9.5%),from$61.54 to $67.41,inthe comparison group,arelativedifference of $1.30PMPM ( P =0.013). Theincreasesinutilization fromthe pre-changeperiodtothe post-changeperiodwerenot significantlydifferent in theintervention group (2.4%) versus the...

Pattern and predictors of the initiation of biologic agents for the treatment of rheumatoid arthritis in the United States: An analysis using a large observational data bank

“…Given the increased cost-shifting to consumers by health care plans, 29 the large price differential between biologics and nonbiologic DMARDs, and the out-of-pocket costs to patients, it is not unexpected that income would be a significant predictor of the initiation of biologics.…”

A Randomized Controlled Trial to Assess Pharmacist-Physician Collaborative Practice in the Management of Metabolic Syndrome in a University Medical Clinic in Jordan

“…27,28 First, baseline values were subtracted from 6-month follow-up values to calculate change amounts. Then, the statistical significance levels of by-group differences in the change amounts were calculated using Student's t-tests for variables with normal distributions and Mann-Whitney U tests when normality and equality of variance assumptions were not met.…”

Making the World Safe for Evidence-Based Policy: Let's Slay the Biases in Research on Value-Based Insurance Design

“…[10][11][12][13][14] Total prescription claims count is a particularly useful measure, predicting future health care expenditures more accurately than did several commonly used comorbidity scales in a 2006 study of antihypertensive medication users. 15 Additional important baseline information includes plan design features (e.g., preferred provider organization vs. health maintenance organization, fee-for-service vs. provider-risk arrangements), geographic region (shown to be a key prescription drug utilization factor in previous reports), 16 industry or sector (e.g., white collar vs. blue collar) especially if just 2 employers are being compared (i.e., as opposed to comparing study groups comprising numerous small employers), and other relevant pharmacy benefit design features (e.g., step therapy, mail order option).…”