- Email: [email protected]
TR3 TIME-RELEASE LAUNCH IN RELATION TO GENERIC LAUNCH: CASES
A8
Abstracts
estimates of the effects of individual ACEIs on mortality. Multiple PS may be used more often if no information needed to predict outcomes is lost from sub sampling. SB3 COMPARISON OF DIFFERENT PROPENSITY SCORE MATCHING METHODS IN ELDERLY ANTIPSYCHOTIC USERS Mehta S, Chen H, Johnson M, Aparasu RR University of Houston, Houston, TX, USA OBJECTIVES: Various propensity score matching techniques are used in observational studies to reduce selection and confounding bias. The purpose of this study was to evaluate the four most popularly used matching methods namely Mahalanobis metric matching within calipers of propensity scores, caliper matching with and without replacement and greedy matching for making elderly antipsychotic users comparable. METHODS: IMS LifeLink™ Claims were utilized to identify elderly patients using atypical and typical antipsychotics. Eighty covariates including demographics, hospitalization, co-morbidities and co-medications were used to match typical and atypical antipsychotic users using propensity scores matching. Propensity matching methods were evaluated on the basis of following criteria: (1) Number of variables which remains significant after matching using t-test and chi-square; (2) Percentage bias reduction for the variables which remained significant after matching; (3) Mean difference in propensity scores as a percentage of average standard deviation (SD); and (4) Density estimates of the propensity scores of the two groups using the Kolmogorov-Smirnov test. RESULTS: The four matching methods reduced bias by making two groups comparable. However greedy matching yielded the best results when the four criteria were applied. Only 5 explanatory variables remained significant after greedy matching compared to 36, 43 and 9 with Mahalanobis metric matching, and caliper matching with and without replacement, respectively. More than 90% bias reduction was obtained through all the matching methods. Mean difference as a percentage of the average SD was 0% with greedy and caliper matching with replacement and these were the only techniques that produced propensity scores densities with insignificant differences. CONCLUSIONS: The greedy matching technique was found to be efficient in matching different classes of antipsychotic users. Although the efficiency of matching methods could differ based on the study sample and availability of covariates, a priori criteria can be useful in selecting the most appropriate matching technique. SB4 DEALING WITH SELECTION BIAS IN NONLINEAR SETTINGS: A CASE OF COMPARATIVE EFFECTIVENESS OF STATIN PLUS FIBRATE COMBINATION THERAPY VERSUS STATIN MONOTHERAPY IN TYPE II DIABETES Suh HS1, Doctor J2, Hay JW3 1 National Evidence-based Healthcare Collaborating Agency (NECA), Seoul, South Korea, 2 USC School of Pharmacy, Los Angeles, CA, USA, 3University of Southern California, Los Angeles, CA, USA OBJECTIVES: To estimate the effectiveness of statin+fibrate combination-therapy versus statin-monotherapy on cardiovascular disease(CVD) occurrence in subjects with type II diabetes in a managed care setting using appropriate econometric models dealing with selection bias in nonlinear settings. METHODS: Combination-therapygroup and monotherapy-group were identified among subjects with type II diabetes with two-years intake period(7/1/2002–6/30/2004) and three-years follow-up using administrative claims from a US health plan covering four million lives. Outcomes measure was CVD-occurrence. A univariate-probit model was developed to evaluate adjusted CVD-risk difference between groups. To control for selection bias, we used propensity score(PS) and instrumental variable(IV) method. To deal with nonlinear outcomes, we built two-stage-probit model with IV method using two-stage-residualinclusion estimation. We used physician prescribing preference as the instrument. To test the validity of the instrument, we tested for the correlation between the instrument and treatment indicator using standard t-test. To check whether it is valid to exclude the instrument from the main equation, Wald-test was performed. Stock-and-Yogo test was used to check the weak instrument issue. To test the endogeneity of treatment indicator, we performed Hausman-test. RESULTS: Adjusting for age, gender, priorCVD, CVD-related pharmacy-costs, Elixhauser-comorbidity, and diabetes with complication, combination-therapy-group experienced 9.1% less CVD compared with monotherapy-group at the mean of covariates(P = 0.008). The results from probit and PS model were similar. For the IV model, specification-tests indicated that the validity of the instrument was satisfied. However, Hausman-test implied that treatment-indicator was not endogenous(p = 0.172). CONCLUSIONS: To deal with nonlinearity issues when using IV method, we employed two-stage residual-inclusion estimation. Since we failed to identify selection bias which may be due to untestable assumptions and treatment effect heterogeneity, a univariate-probit model or PS model was used to draw conclusions. In diabetics after adjusting for known baseline differences, CVDrisk was significantly lower among subjects with statin+fibrate combination-therapy compared with those with statin-monotherapy.
PODIUM SESSION II: HEALTH CARE TRENDS TR1 THE EFFECT OF GENERIC DRUG ENTRY ON U.S. MEDICAID EXPENDITURES: 1991–2008 Kelton CM1, Guo JJ2, Safi A1, Yu Y1 1 University of Cincinnati College of Business, Cincinnati, OH, USA, 2University of Cincinnati Medical Center, Cincinnati, OH, USA OBJECTIVES: After the patent expires for a branded pharmaceutical, generic drug companies sell bioequivalent versions of the drug, leading to lower prices and reduced expenditures for payers. The objectives of this study were to 1) predict the number of generic-company entrants in the market, and, 2) determine the trend in drug price post-entry based on drug, market, and firm characteristics. METHODS: Using the national summary file of Medicaid outpatient drug utilization maintained by the Centers for Medicare and Medicaid Services, quarterly utilization and expenditure data from 1991–2008 were extracted for 40 drugs that experienced initial generic entry between 1992 and 2004. Generic relative price (GRP) was constructed as reimbursement per unit for a specific firm and quarter divided by average reimbursement per unit over the year before entry. Least-squares regression models were estimated on the panel data to explain number of entrants, GRP, and average GRP across firms (AGRP). RESULTS: After patent expiration, the number of generic-firm entrants ranged from 1 to 25. Significant (p < 0.0001) predictors for number of entrants included pre-entry market size, market growth, number of quarters since entry, and administration form (oral, injectable, or topical). The number of firms had a statistically significant (p < 0.0001), nonlinear negative effect on GRP and AGRP. With the addition of one more generic firm, GRP is expected to fall by 0.018; AGRP is expected to fall by 0.053. High demand, as indicated by high post-entry expenditures, had a statistically significant (p < 0.0001) positive effect on both GRP and AGRP. Many of the firm-specific effects were significant (p < 0.0001). CONCLUSIONS: Medicaid can generally look forward to cost savings following generic entry. However, a small initial market size or a drop in demand following entry prohibits Medicaid from obtaining cost relief. Differences in pricing strategies across firms were indicated by the variation in GRP across manufacturers. TR2 RECENT TRENDS IN EMPLOYER HEALTH CARE SPENDING BY DISEASE Huse DM, Marder WD Thomson Reuters (Healthcare), Cambridge, MA, USA OBJECTIVES: Efforts to “bend the cost curve” in US health care need to be informed by a differentiated view of health care spending by disease. We highlight the importance of particular diseases based on level and rate of growth of spending. METHODS: National health care spending by employers for all services and for prescription drugs was estimated using claims data from employer sponsored health plans from 2004 and 2008. Data were projected using sampling weights derived from the Medical Expenditure Panel Survey and trends in spending were calculated as compound annual growth rates (CAGR). Claims were assigned to disease-specific episodes using the Thomson Reuters Medical Episode Grouper (MEG). RESULTS: Annual spending data were analyzed for 122 employers with 11.5 million covered lives in 2004 and 143 employers with 15.6 million covered lives in 2008. Overall spending grew at 8.6% per annum while prescription drug spending grew at 10.8%. Musculoskeletal conditions were the most important body system, representing 18% of all spending and growing at 11% per year, well above average. Several specific conditions had total spending growth rates of 15%, including osteoarthritis of the spine, multiple sclerosis, and prostate cancer. Preventive health services also grew at 15%. The single disease with the largest total spending, angina pectoris, grew at less than 1% per year. Drug spending grew the fastest for Crohn’s disease (25%), epilepsy (23%), and multiple sclerosis (19%). Spending for type 2 diabetes drugs was particularly important, as it represented 11% of all drug spending and grew at 17% per year, well above average. CONCLUSIONS: Major drivers of recent growth in employer health care spending are conditions related to obesity, including musculoskeletal and endocrine conditions. The highest rates of drug spending were observed in conditions where biologic therapies are becoming standard of care. TR3 TIME-RELEASE LAUNCH IN RELATION TO GENERIC LAUNCH: CASES Wiederkehr DP, Nichols E, Doyle J Quintiles Global Consulting, Hawthorne, NY, USA OBJECTIVES: While generic drug use has become more common with the availability of generic alternatives to the many high volume branded products, the rate at which generic drugs have displaced branded counterparts is not well understood, particularly when time-release (TR) brands are introduced within a similar timeframe. We examined the uptake of generic drugs with respect to products and classes in conjunction with the uptake of time-release formulations of branded products to determine the impact of generic uptake on branded drugs, particularly when a branded time-release formulation is launched prior to generic competition. METHODS: SSRIs, atypical antipsychotics, nonbenzodiazepines as well as azythromycin were selected as case classes and products. From 1995 to 2009 the volume of prescriptions were collected monthly using SDI’s VONA databases and grouped according to branded and generic sales by active molecule. RESULTS: Among SSRIs, brands that launched TR branded products at least one-year prior to the launch of the original formulation generic, such as paroxetine, saw less erosion of the total brand than those with TR launch less than one year or no TR launch. Similarly, among nonbenzodiazepines, early launch of TR
Abstracts formulation resulted in less erosion from the total brand. In cases where branded utilization is not converted to TR formulation prior to launch of generic competition, such as risperidone and fluoxetine, or TR launch precedes generic launch by less than six months or not at all, such as azythromycin and sertraline, dramatic reduction in total sales brand was observed. In the unique case where branded escitalopram was introduced prior to the generic launch of citalopram, an effect similar to a branded TR launch was observed. CONCLUSIONS: While generic use typically replaces the majority of branded equivalents in a short timeframe, branded products with a timerelease formulation may limit uptake rate of the generic version of the total branded product. TR4 PUBLIC HEALTH AND ECONOMIC IMPACT OF 13-VALENT PNEUMOCOCCAL CONJUGATE VACCINE (PCV13) IN AN INFLUENZA PANDEMIC IN SINGAPORE AND HONG KONG Rubin J1, McGarry L1, Klugman K2, Strutton D3, Gilmore K1, Hwang S3, Rinaldi F4, Weinstein M5 1 i3 Innovus, Medford, MA, USA, 2Rollins School of Public Health, Emory University, Atlanta, GA, USA, 3Pfizer Inc., Collegeville, PA, USA, 4Independent Consultant to Wyeth—Asia Pacific, London, UK, 5Harvard School of Public Health, Boston, MA, USA OBJECTIVES: Historic data suggest that most 1918 influenza pandemic-related deaths were due to pneumococcal disease (PD); preliminary evidence shows a similar pattern for the 2009 H1N1 influenza outbreak. Implementation of 13-valent pneumococcal conjugate vaccine (PCV13) national immunization programs would likely lessen the impact of a pandemic in Asia-Pacific populations that currently have low pneumococcal vaccination rates; our objective was to quantify this impact in Singapore and Hong Kong. METHODS: We used a decision-analytic model to assess the impact of PCV13 infant vaccination on PD incidence and mortality in an influenza pandemic in Singapore and Hong Kong versus no vaccination. The model was estimated from published sources; both direct and indirect (herd) effects against PD were included. Effectiveness of PCV13 was extrapolated from observed US 7-valent PCV (PCV7) data, using assumptions on serotype prevalence in the pre-PCV7 era, and PCV13 protection against the 6 serotypes not in PCV7. Country-specific data were used where available; where unavailable, US data were used. To simulate acceleration of PD transmission in a flu pandemic, we calibrated to 1918 estimates of countryspecific incidence and mortality, adjusting mortality for the advent of antibiotics. PD incidence and mortality, and total PD-related health-care costs were evaluated over a 1-year horizon. Results are reported in local 2008 currency. RESULTS: Preliminary results of the model indicate that in a pandemic of 1918 severity, PCV13 vaccination would prevent 3,300 cases of hospitalized pneumococcal pneumonia and 320 deaths in Singapore, and save SGD1.53 million in health-care costs (net of vaccination). In Hong Kong, PCV13 would prevent 8,200 cases of hospitalized pneumonia (all-cause), and 2,200 pneumonia deaths, respectively; PCV13 would save HKD250 million in medical costs. CONCLUSIONS: In an influenza pandemic affecting the Asia-Pacific region, infant vaccination with PCV13 would likely be highly effective in reducing pandemic-related deaths, PD cases and associated costs.
PODIUM SESSION III: HEALTH CARE DECISION-MAKER’S CASE STUDIES CASE1 IMPLEMENTATION OF A NONFORMULARY PRESCRIBING MONITORING SYSTEMS REVEALS OPPORTUNITIES FOR COST SAVINGS IN AN ACADEMIC MEDICAL CENTER Helmons PJ, Daniels CE University of California San Diego Medical Center, San Diego, CA, USA ORGANIZATION: University of California San Diego Medical Center (UCSDMC) is a 548 bed academic medical center with an average daily census of 368 patients and 23,000 annual discharges. UCSDMC has a large regional footprint as it is the only academic medical center in the region and serves as the regions burn center and high risk obstetrics and neonatal care center. In addition, UCSDMC was the regions first level 1 trauma unit. Consequently, patients are admitted to UCSDMC with a wide variety of medications and insurance coverage. From a medication safety standpoint, outpatient maintenance medication is generally not substituted in our hospital. In addition, patients are discouraged to bring their own medication into the hospital to assure the quality of the medications. These patient population and policy characteristics make formulary management particularly challenging, as this practice results in a constant baseline use of nonformulary outpatient maintenance medications in our inpatient patient population. PROBLEM OR ISSUE ADDRESSED: Detecting trends in nonformulary prescribing is important as increased use of nonformulary medications could indicate an important clinical need of this medication. On the other hand, increased use of nonformulary medications in therapeutic areas with equivalent but less expensive formulary medications, leads to increased costs. We hypothesized that analyzing trends in nonformulary prescribing leads to detection of irrational and more expensive, nonformulary medications and therefore to cost savings. GOALS: To address these issues, we developed a nonformulary medication use monitoring system designed to improve formulary adherence by 50% and detect irrational nonformulary use at an early stage. OUTCOMES ITEMS USED IN THE DECISION: Formulary adherence is defined as the number of nonformulary medication doses billed divided by the total number of doses billed. As an example of irrational prescribing detected
A9 by this system, cost savings associated with reversing nonformulary levalbuterol use to formulary albuterol are reported. IMPLEMENTATION STRATEGY: Trends in nonformulary use are monitored using daily, monthly and semi-annual reports.— Daily monitoring is done by the clinical pharmacist. A report that specifies the nonformulary orders per nursing unit is sent to each pharmacist for follow up. The pharmacist evaluates the order for appropriateness and contacts the physician if substitution to a formulary alternative is possible. The intervention is recorded in the pharmacy information system and compliance is monitored by the pharmacoeconomics specialist.—Monthly monitoring is done by the pharmacoeconomics specialist. The results of this monthly analysis are discussed with the Pharmacist-in-Chief and with the Chair of the P&T Committee.—Semi-annual reporting is done by the pharmacoeconomics specialist. Trends in nonformulary prescribing over a six month period are discussed at the P&T meeting. An action plan is approved to address the findings. RESULTS: The first semi-annual nonformulary prescribing report showed a nonformulary use of 0.26% of billed doses. The most frequently prescribed nonformulary medications were: levalbuterol (bronchodilating agent), latanoprost (ocular glaucoma agent), pentosan (bladder relief agent) and Mg-Al plus (magnesium supplement). Based on this initial report and the lack of formulary alternatives, the P&T committee added the latter three agents to the formulary. However, from a cost savings standpoint, the use of levalbuterol at our institution was discouraged as this agent has no known benefits over our formulary agent albuterol. At the time of the report, levalbuterol was almost 4 times more expensive than albuterol. Collaboration with respiratory therapy and pulmonology was sought to educate prescribers and pharmacists about the therapeutic equivalence of albuterol and levalbuterol. After these interventions, the second semi-annual report showed an 80% decrease of nonformulary use to 0.05% of billed doses. Levalbuterol use decreased by 67% from an average of 24 patients per month in the 6 month period before the intervention to 8 patients after the intervention. Levalbuterol expenditure decreased from $7800 to $1421 over the same period and continues to drop, corresponding to an annualized cost avoidance of $8000. LESSONS LEARNED: This case study shows that in an environment where formulary management is challenging, a nonformulary prescribing monitoring system is successful in decreasing nonformulary prescribing and identifying areas of improvement. The large decrease in use of levalbuterol is only one of the successes resulting from this effort. However, by implementing continuous monitoring of nonformulary prescribing we expect to detect and prevent these excesses in the future. We acknowledge that the monetary savings of decreasing levalbuterol use and nonformulary prescribing as a whole is marginal. However, other benefits of a formulary that is better tailored to the institution’s needs could include better inventory management, improved medication safety and less interruptions of therapy. This will be the focus of a subsequent study. CASE2 INVENTING A TIERED FORMULARY BASED ON THE INCREMENTAL COST-EFFECTIVENESS OF DRUGS Watkins J1, Sullivan S2, Ramsey S3, Wong E1, Veenstra D2, Burke W4, Roy Choudhury S1 1 Premera Blue Cross, Mountlake Terrace, WA, USA, 2University of Washington, Pharmaceutical Outcomes Research and Policy Program, Seattle, WA, USA, 3Fred Hutchinson Cancer Research Center, University of Washington, Seattle, WA, USA, 4 University of Washington, Department of Bioethics and Humanities, Seattle, WA, USA ORGANIZATION: Premera Blue Cross PROBLEM OR ISSUE ADDRESSED: Traditional U.S. payer drug selection processes are based on safety, efficacy and drug cost. As premiums continue to rise, many plan sponsors have responded by raising member cost shares on biologics and other high cost pharmaceuticals. This creates affordability barriers for these drugs, regardless of their clinical value, comparative effectiveness or cost-effectiveness. GOALS: Our novel cost-effectiveness driven formulary design seeks to offset the negative impact of such policies on clinical outcomes and quality of care by assigning the drugs that are the most incrementally cost-effective to the lowest formulary tiers. OUTCOMES ITEMS USED IN THE DECISION: Comparative effectiveness and safety (clinical benefits and harms as determined by our existing formulary process) and incremental cost-effectiveness ratios of specific therapies vs. usual care alternatives, based on published cost-effectiveness analyses, manufacturer-submitted cost-effectiveness models and in some cases, models developed internally by plan staff. Cost-effectiveness studies are subjected to a quality review process before being presented to the formulary committee. IMPLEMENTATION STRATEGY: Premera Blue Cross has created a formulary process that assigns each drug to one of four tiers based on a set of flexible cost-effectiveness thresholds. Incremental costeffectiveness, assessed as rigorously as possible, provides a scientific, transparent and defensible means of maintaining lower cost shares for the highest value drugs, regardless of actual prescription cost. In some cases, a drug may fall in more than one tier, based on the patient population and the particular indication for which it is being used. This presentation describes Premera’s formulary process, how the thresholds for each tier were established and how drugs and therapeutic classes are reviewed by an external multidisciplinary expert committee, with specific therapeutic drug class outcome examples. This formulary is being offered as part of a new pharmacy benefit product addressing a currently unmet market need. RESULTS: Premera’s new formulary process was developed in 2009. This presentation describes the methods used to establish incremental cost-effectiveness thresholds for each formulary tier and the constitution of an external multidisciplinary expert committee to assign specific drugs to tiers. The rationale and tier assignments for specific therapeutic drug classes are presented as examples. LESSONS LEARNED: Despite limitations to the current availability of clinical and economic outcomes from well-designed studies on which to base tiering decisions, it is possible to implement a credible process based on incremental
Abstracts
estimates of the effects of individual ACEIs on mortality. Multiple PS may be used more often if no information needed to predict outcomes is lost from sub sampling. SB3 COMPARISON OF DIFFERENT PROPENSITY SCORE MATCHING METHODS IN ELDERLY ANTIPSYCHOTIC USERS Mehta S, Chen H, Johnson M, Aparasu RR University of Houston, Houston, TX, USA OBJECTIVES: Various propensity score matching techniques are used in observational studies to reduce selection and confounding bias. The purpose of this study was to evaluate the four most popularly used matching methods namely Mahalanobis metric matching within calipers of propensity scores, caliper matching with and without replacement and greedy matching for making elderly antipsychotic users comparable. METHODS: IMS LifeLink™ Claims were utilized to identify elderly patients using atypical and typical antipsychotics. Eighty covariates including demographics, hospitalization, co-morbidities and co-medications were used to match typical and atypical antipsychotic users using propensity scores matching. Propensity matching methods were evaluated on the basis of following criteria: (1) Number of variables which remains significant after matching using t-test and chi-square; (2) Percentage bias reduction for the variables which remained significant after matching; (3) Mean difference in propensity scores as a percentage of average standard deviation (SD); and (4) Density estimates of the propensity scores of the two groups using the Kolmogorov-Smirnov test. RESULTS: The four matching methods reduced bias by making two groups comparable. However greedy matching yielded the best results when the four criteria were applied. Only 5 explanatory variables remained significant after greedy matching compared to 36, 43 and 9 with Mahalanobis metric matching, and caliper matching with and without replacement, respectively. More than 90% bias reduction was obtained through all the matching methods. Mean difference as a percentage of the average SD was 0% with greedy and caliper matching with replacement and these were the only techniques that produced propensity scores densities with insignificant differences. CONCLUSIONS: The greedy matching technique was found to be efficient in matching different classes of antipsychotic users. Although the efficiency of matching methods could differ based on the study sample and availability of covariates, a priori criteria can be useful in selecting the most appropriate matching technique. SB4 DEALING WITH SELECTION BIAS IN NONLINEAR SETTINGS: A CASE OF COMPARATIVE EFFECTIVENESS OF STATIN PLUS FIBRATE COMBINATION THERAPY VERSUS STATIN MONOTHERAPY IN TYPE II DIABETES Suh HS1, Doctor J2, Hay JW3 1 National Evidence-based Healthcare Collaborating Agency (NECA), Seoul, South Korea, 2 USC School of Pharmacy, Los Angeles, CA, USA, 3University of Southern California, Los Angeles, CA, USA OBJECTIVES: To estimate the effectiveness of statin+fibrate combination-therapy versus statin-monotherapy on cardiovascular disease(CVD) occurrence in subjects with type II diabetes in a managed care setting using appropriate econometric models dealing with selection bias in nonlinear settings. METHODS: Combination-therapygroup and monotherapy-group were identified among subjects with type II diabetes with two-years intake period(7/1/2002–6/30/2004) and three-years follow-up using administrative claims from a US health plan covering four million lives. Outcomes measure was CVD-occurrence. A univariate-probit model was developed to evaluate adjusted CVD-risk difference between groups. To control for selection bias, we used propensity score(PS) and instrumental variable(IV) method. To deal with nonlinear outcomes, we built two-stage-probit model with IV method using two-stage-residualinclusion estimation. We used physician prescribing preference as the instrument. To test the validity of the instrument, we tested for the correlation between the instrument and treatment indicator using standard t-test. To check whether it is valid to exclude the instrument from the main equation, Wald-test was performed. Stock-and-Yogo test was used to check the weak instrument issue. To test the endogeneity of treatment indicator, we performed Hausman-test. RESULTS: Adjusting for age, gender, priorCVD, CVD-related pharmacy-costs, Elixhauser-comorbidity, and diabetes with complication, combination-therapy-group experienced 9.1% less CVD compared with monotherapy-group at the mean of covariates(P = 0.008). The results from probit and PS model were similar. For the IV model, specification-tests indicated that the validity of the instrument was satisfied. However, Hausman-test implied that treatment-indicator was not endogenous(p = 0.172). CONCLUSIONS: To deal with nonlinearity issues when using IV method, we employed two-stage residual-inclusion estimation. Since we failed to identify selection bias which may be due to untestable assumptions and treatment effect heterogeneity, a univariate-probit model or PS model was used to draw conclusions. In diabetics after adjusting for known baseline differences, CVDrisk was significantly lower among subjects with statin+fibrate combination-therapy compared with those with statin-monotherapy.
PODIUM SESSION II: HEALTH CARE TRENDS TR1 THE EFFECT OF GENERIC DRUG ENTRY ON U.S. MEDICAID EXPENDITURES: 1991–2008 Kelton CM1, Guo JJ2, Safi A1, Yu Y1 1 University of Cincinnati College of Business, Cincinnati, OH, USA, 2University of Cincinnati Medical Center, Cincinnati, OH, USA OBJECTIVES: After the patent expires for a branded pharmaceutical, generic drug companies sell bioequivalent versions of the drug, leading to lower prices and reduced expenditures for payers. The objectives of this study were to 1) predict the number of generic-company entrants in the market, and, 2) determine the trend in drug price post-entry based on drug, market, and firm characteristics. METHODS: Using the national summary file of Medicaid outpatient drug utilization maintained by the Centers for Medicare and Medicaid Services, quarterly utilization and expenditure data from 1991–2008 were extracted for 40 drugs that experienced initial generic entry between 1992 and 2004. Generic relative price (GRP) was constructed as reimbursement per unit for a specific firm and quarter divided by average reimbursement per unit over the year before entry. Least-squares regression models were estimated on the panel data to explain number of entrants, GRP, and average GRP across firms (AGRP). RESULTS: After patent expiration, the number of generic-firm entrants ranged from 1 to 25. Significant (p < 0.0001) predictors for number of entrants included pre-entry market size, market growth, number of quarters since entry, and administration form (oral, injectable, or topical). The number of firms had a statistically significant (p < 0.0001), nonlinear negative effect on GRP and AGRP. With the addition of one more generic firm, GRP is expected to fall by 0.018; AGRP is expected to fall by 0.053. High demand, as indicated by high post-entry expenditures, had a statistically significant (p < 0.0001) positive effect on both GRP and AGRP. Many of the firm-specific effects were significant (p < 0.0001). CONCLUSIONS: Medicaid can generally look forward to cost savings following generic entry. However, a small initial market size or a drop in demand following entry prohibits Medicaid from obtaining cost relief. Differences in pricing strategies across firms were indicated by the variation in GRP across manufacturers. TR2 RECENT TRENDS IN EMPLOYER HEALTH CARE SPENDING BY DISEASE Huse DM, Marder WD Thomson Reuters (Healthcare), Cambridge, MA, USA OBJECTIVES: Efforts to “bend the cost curve” in US health care need to be informed by a differentiated view of health care spending by disease. We highlight the importance of particular diseases based on level and rate of growth of spending. METHODS: National health care spending by employers for all services and for prescription drugs was estimated using claims data from employer sponsored health plans from 2004 and 2008. Data were projected using sampling weights derived from the Medical Expenditure Panel Survey and trends in spending were calculated as compound annual growth rates (CAGR). Claims were assigned to disease-specific episodes using the Thomson Reuters Medical Episode Grouper (MEG). RESULTS: Annual spending data were analyzed for 122 employers with 11.5 million covered lives in 2004 and 143 employers with 15.6 million covered lives in 2008. Overall spending grew at 8.6% per annum while prescription drug spending grew at 10.8%. Musculoskeletal conditions were the most important body system, representing 18% of all spending and growing at 11% per year, well above average. Several specific conditions had total spending growth rates of 15%, including osteoarthritis of the spine, multiple sclerosis, and prostate cancer. Preventive health services also grew at 15%. The single disease with the largest total spending, angina pectoris, grew at less than 1% per year. Drug spending grew the fastest for Crohn’s disease (25%), epilepsy (23%), and multiple sclerosis (19%). Spending for type 2 diabetes drugs was particularly important, as it represented 11% of all drug spending and grew at 17% per year, well above average. CONCLUSIONS: Major drivers of recent growth in employer health care spending are conditions related to obesity, including musculoskeletal and endocrine conditions. The highest rates of drug spending were observed in conditions where biologic therapies are becoming standard of care. TR3 TIME-RELEASE LAUNCH IN RELATION TO GENERIC LAUNCH: CASES Wiederkehr DP, Nichols E, Doyle J Quintiles Global Consulting, Hawthorne, NY, USA OBJECTIVES: While generic drug use has become more common with the availability of generic alternatives to the many high volume branded products, the rate at which generic drugs have displaced branded counterparts is not well understood, particularly when time-release (TR) brands are introduced within a similar timeframe. We examined the uptake of generic drugs with respect to products and classes in conjunction with the uptake of time-release formulations of branded products to determine the impact of generic uptake on branded drugs, particularly when a branded time-release formulation is launched prior to generic competition. METHODS: SSRIs, atypical antipsychotics, nonbenzodiazepines as well as azythromycin were selected as case classes and products. From 1995 to 2009 the volume of prescriptions were collected monthly using SDI’s VONA databases and grouped according to branded and generic sales by active molecule. RESULTS: Among SSRIs, brands that launched TR branded products at least one-year prior to the launch of the original formulation generic, such as paroxetine, saw less erosion of the total brand than those with TR launch less than one year or no TR launch. Similarly, among nonbenzodiazepines, early launch of TR
Abstracts formulation resulted in less erosion from the total brand. In cases where branded utilization is not converted to TR formulation prior to launch of generic competition, such as risperidone and fluoxetine, or TR launch precedes generic launch by less than six months or not at all, such as azythromycin and sertraline, dramatic reduction in total sales brand was observed. In the unique case where branded escitalopram was introduced prior to the generic launch of citalopram, an effect similar to a branded TR launch was observed. CONCLUSIONS: While generic use typically replaces the majority of branded equivalents in a short timeframe, branded products with a timerelease formulation may limit uptake rate of the generic version of the total branded product. TR4 PUBLIC HEALTH AND ECONOMIC IMPACT OF 13-VALENT PNEUMOCOCCAL CONJUGATE VACCINE (PCV13) IN AN INFLUENZA PANDEMIC IN SINGAPORE AND HONG KONG Rubin J1, McGarry L1, Klugman K2, Strutton D3, Gilmore K1, Hwang S3, Rinaldi F4, Weinstein M5 1 i3 Innovus, Medford, MA, USA, 2Rollins School of Public Health, Emory University, Atlanta, GA, USA, 3Pfizer Inc., Collegeville, PA, USA, 4Independent Consultant to Wyeth—Asia Pacific, London, UK, 5Harvard School of Public Health, Boston, MA, USA OBJECTIVES: Historic data suggest that most 1918 influenza pandemic-related deaths were due to pneumococcal disease (PD); preliminary evidence shows a similar pattern for the 2009 H1N1 influenza outbreak. Implementation of 13-valent pneumococcal conjugate vaccine (PCV13) national immunization programs would likely lessen the impact of a pandemic in Asia-Pacific populations that currently have low pneumococcal vaccination rates; our objective was to quantify this impact in Singapore and Hong Kong. METHODS: We used a decision-analytic model to assess the impact of PCV13 infant vaccination on PD incidence and mortality in an influenza pandemic in Singapore and Hong Kong versus no vaccination. The model was estimated from published sources; both direct and indirect (herd) effects against PD were included. Effectiveness of PCV13 was extrapolated from observed US 7-valent PCV (PCV7) data, using assumptions on serotype prevalence in the pre-PCV7 era, and PCV13 protection against the 6 serotypes not in PCV7. Country-specific data were used where available; where unavailable, US data were used. To simulate acceleration of PD transmission in a flu pandemic, we calibrated to 1918 estimates of countryspecific incidence and mortality, adjusting mortality for the advent of antibiotics. PD incidence and mortality, and total PD-related health-care costs were evaluated over a 1-year horizon. Results are reported in local 2008 currency. RESULTS: Preliminary results of the model indicate that in a pandemic of 1918 severity, PCV13 vaccination would prevent 3,300 cases of hospitalized pneumococcal pneumonia and 320 deaths in Singapore, and save SGD1.53 million in health-care costs (net of vaccination). In Hong Kong, PCV13 would prevent 8,200 cases of hospitalized pneumonia (all-cause), and 2,200 pneumonia deaths, respectively; PCV13 would save HKD250 million in medical costs. CONCLUSIONS: In an influenza pandemic affecting the Asia-Pacific region, infant vaccination with PCV13 would likely be highly effective in reducing pandemic-related deaths, PD cases and associated costs.
PODIUM SESSION III: HEALTH CARE DECISION-MAKER’S CASE STUDIES CASE1 IMPLEMENTATION OF A NONFORMULARY PRESCRIBING MONITORING SYSTEMS REVEALS OPPORTUNITIES FOR COST SAVINGS IN AN ACADEMIC MEDICAL CENTER Helmons PJ, Daniels CE University of California San Diego Medical Center, San Diego, CA, USA ORGANIZATION: University of California San Diego Medical Center (UCSDMC) is a 548 bed academic medical center with an average daily census of 368 patients and 23,000 annual discharges. UCSDMC has a large regional footprint as it is the only academic medical center in the region and serves as the regions burn center and high risk obstetrics and neonatal care center. In addition, UCSDMC was the regions first level 1 trauma unit. Consequently, patients are admitted to UCSDMC with a wide variety of medications and insurance coverage. From a medication safety standpoint, outpatient maintenance medication is generally not substituted in our hospital. In addition, patients are discouraged to bring their own medication into the hospital to assure the quality of the medications. These patient population and policy characteristics make formulary management particularly challenging, as this practice results in a constant baseline use of nonformulary outpatient maintenance medications in our inpatient patient population. PROBLEM OR ISSUE ADDRESSED: Detecting trends in nonformulary prescribing is important as increased use of nonformulary medications could indicate an important clinical need of this medication. On the other hand, increased use of nonformulary medications in therapeutic areas with equivalent but less expensive formulary medications, leads to increased costs. We hypothesized that analyzing trends in nonformulary prescribing leads to detection of irrational and more expensive, nonformulary medications and therefore to cost savings. GOALS: To address these issues, we developed a nonformulary medication use monitoring system designed to improve formulary adherence by 50% and detect irrational nonformulary use at an early stage. OUTCOMES ITEMS USED IN THE DECISION: Formulary adherence is defined as the number of nonformulary medication doses billed divided by the total number of doses billed. As an example of irrational prescribing detected
A9 by this system, cost savings associated with reversing nonformulary levalbuterol use to formulary albuterol are reported. IMPLEMENTATION STRATEGY: Trends in nonformulary use are monitored using daily, monthly and semi-annual reports.— Daily monitoring is done by the clinical pharmacist. A report that specifies the nonformulary orders per nursing unit is sent to each pharmacist for follow up. The pharmacist evaluates the order for appropriateness and contacts the physician if substitution to a formulary alternative is possible. The intervention is recorded in the pharmacy information system and compliance is monitored by the pharmacoeconomics specialist.—Monthly monitoring is done by the pharmacoeconomics specialist. The results of this monthly analysis are discussed with the Pharmacist-in-Chief and with the Chair of the P&T Committee.—Semi-annual reporting is done by the pharmacoeconomics specialist. Trends in nonformulary prescribing over a six month period are discussed at the P&T meeting. An action plan is approved to address the findings. RESULTS: The first semi-annual nonformulary prescribing report showed a nonformulary use of 0.26% of billed doses. The most frequently prescribed nonformulary medications were: levalbuterol (bronchodilating agent), latanoprost (ocular glaucoma agent), pentosan (bladder relief agent) and Mg-Al plus (magnesium supplement). Based on this initial report and the lack of formulary alternatives, the P&T committee added the latter three agents to the formulary. However, from a cost savings standpoint, the use of levalbuterol at our institution was discouraged as this agent has no known benefits over our formulary agent albuterol. At the time of the report, levalbuterol was almost 4 times more expensive than albuterol. Collaboration with respiratory therapy and pulmonology was sought to educate prescribers and pharmacists about the therapeutic equivalence of albuterol and levalbuterol. After these interventions, the second semi-annual report showed an 80% decrease of nonformulary use to 0.05% of billed doses. Levalbuterol use decreased by 67% from an average of 24 patients per month in the 6 month period before the intervention to 8 patients after the intervention. Levalbuterol expenditure decreased from $7800 to $1421 over the same period and continues to drop, corresponding to an annualized cost avoidance of $8000. LESSONS LEARNED: This case study shows that in an environment where formulary management is challenging, a nonformulary prescribing monitoring system is successful in decreasing nonformulary prescribing and identifying areas of improvement. The large decrease in use of levalbuterol is only one of the successes resulting from this effort. However, by implementing continuous monitoring of nonformulary prescribing we expect to detect and prevent these excesses in the future. We acknowledge that the monetary savings of decreasing levalbuterol use and nonformulary prescribing as a whole is marginal. However, other benefits of a formulary that is better tailored to the institution’s needs could include better inventory management, improved medication safety and less interruptions of therapy. This will be the focus of a subsequent study. CASE2 INVENTING A TIERED FORMULARY BASED ON THE INCREMENTAL COST-EFFECTIVENESS OF DRUGS Watkins J1, Sullivan S2, Ramsey S3, Wong E1, Veenstra D2, Burke W4, Roy Choudhury S1 1 Premera Blue Cross, Mountlake Terrace, WA, USA, 2University of Washington, Pharmaceutical Outcomes Research and Policy Program, Seattle, WA, USA, 3Fred Hutchinson Cancer Research Center, University of Washington, Seattle, WA, USA, 4 University of Washington, Department of Bioethics and Humanities, Seattle, WA, USA ORGANIZATION: Premera Blue Cross PROBLEM OR ISSUE ADDRESSED: Traditional U.S. payer drug selection processes are based on safety, efficacy and drug cost. As premiums continue to rise, many plan sponsors have responded by raising member cost shares on biologics and other high cost pharmaceuticals. This creates affordability barriers for these drugs, regardless of their clinical value, comparative effectiveness or cost-effectiveness. GOALS: Our novel cost-effectiveness driven formulary design seeks to offset the negative impact of such policies on clinical outcomes and quality of care by assigning the drugs that are the most incrementally cost-effective to the lowest formulary tiers. OUTCOMES ITEMS USED IN THE DECISION: Comparative effectiveness and safety (clinical benefits and harms as determined by our existing formulary process) and incremental cost-effectiveness ratios of specific therapies vs. usual care alternatives, based on published cost-effectiveness analyses, manufacturer-submitted cost-effectiveness models and in some cases, models developed internally by plan staff. Cost-effectiveness studies are subjected to a quality review process before being presented to the formulary committee. IMPLEMENTATION STRATEGY: Premera Blue Cross has created a formulary process that assigns each drug to one of four tiers based on a set of flexible cost-effectiveness thresholds. Incremental costeffectiveness, assessed as rigorously as possible, provides a scientific, transparent and defensible means of maintaining lower cost shares for the highest value drugs, regardless of actual prescription cost. In some cases, a drug may fall in more than one tier, based on the patient population and the particular indication for which it is being used. This presentation describes Premera’s formulary process, how the thresholds for each tier were established and how drugs and therapeutic classes are reviewed by an external multidisciplinary expert committee, with specific therapeutic drug class outcome examples. This formulary is being offered as part of a new pharmacy benefit product addressing a currently unmet market need. RESULTS: Premera’s new formulary process was developed in 2009. This presentation describes the methods used to establish incremental cost-effectiveness thresholds for each formulary tier and the constitution of an external multidisciplinary expert committee to assign specific drugs to tiers. The rationale and tier assignments for specific therapeutic drug classes are presented as examples. LESSONS LEARNED: Despite limitations to the current availability of clinical and economic outcomes from well-designed studies on which to base tiering decisions, it is possible to implement a credible process based on incremental