1. Prescription Drug Expenditures and Healthcare Burdens in the Medicaid Population G. Edward Miller, Jessica S. Banthin and Thomas M Selden AHRQ Conference September 9, 2008

2. Health Care Financial Burdens in the Medicaid Population All state Medicaid programs provide RX benefits with no premiums or deductibles and nominal copayments. All state Medicaid programs provide RX benefits with no premiums or deductibles and nominal copayments. 20 percent of non-elderly adult Medicaid enrollees report difficulty affording RX (Cunningham, 2005). 20 percent of non-elderly adult Medicaid enrollees report difficulty affording RX (Cunningham, 2005). Medicaid enrollees are 3X more likely than persons covered by ESI to live in families with high health care financial burdens (Banthin and Bernard, 2006). Medicaid enrollees are 3X more likely than persons covered by ESI to live in families with high health care financial burdens (Banthin and Bernard, 2006).

3. Medicaid Pharmacy Cost Containment Policies By 2004, most states had implemented at least some cost-containment policies: By 2004, most states had implemented at least some cost-containment policies: – Copayments – Quantity limits (number of prescriptions) – Prior authorization – Generic substitution Goal is to reduce costs Goal is to reduce costs May affect access (Cunningham, 2005; Soumerai, 1994). May affect access (Cunningham, 2005; Soumerai, 1994).

4. Data: Medical Expenditure Panel Survey, 2004-05 The MEPS is an annual survey sponsored by Agency for Healthcare Research & Quality Nationally representative household survey consisting of 12,000 households and 33,000 individuals Nationally representative household survey consisting of 12,000 households and 33,000 individuals Includes data on insurance coverage, health care utilization and expenditures, health status, medical conditions, & more Includes data on insurance coverage, health care utilization and expenditures, health status, medical conditions, & more Most accurate source of data on out of pocket spending for medical care Most accurate source of data on out of pocket spending for medical care Released on public use files, tables, statistical briefs: www.meps.ahrq.gov Released on public use files, tables, statistical briefs: www.meps.ahrq.govwww.meps.ahrq.gov

5. Sample of ‘Medicaid Families’ Goal: study the extent to which families covered by Medicaid are at risk of having high health care burdens Goal: study the extent to which families covered by Medicaid are at risk of having high health care burdens Medicaid families: individuals are included only if their entire family was covered by Medicaid/SCHIP for the entire year. Medicaid families: individuals are included only if their entire family was covered by Medicaid/SCHIP for the entire year. Sample includes: Sample includes: – low income parents and their children – non-elderly adults with disabilities Sample excludes low income elderly: Sample excludes low income elderly: – Medicare coverage affects burden – Since 2006, drug coverage through MMA

6. Research Questions What percentage of non-elderly Medicaid enrollees live in families with health care spending burdens in excess of 5% (10%) of disposable family income? What percentage of non-elderly Medicaid enrollees live in families with health care spending burdens in excess of 5% (10%) of disposable family income? What is the contribution of out-of-pocket (OOP) spending for prescription drugs to overall health care burdens? What is the contribution of out-of-pocket (OOP) spending for prescription drugs to overall health care burdens? Are cost containment policies associated with: Are cost containment policies associated with: – higher OOP spending for drugs? – greater level of financial burdens?

7. Method of Calculating Health Care Financial Burdens Numerator: total out of pocket spending across all individuals in the family. Numerator: total out of pocket spending across all individuals in the family. Denominator: total family income and adjusted for taxes. Denominator: total family income and adjusted for taxes. We identify individuals living in families that spend more than 5% or more than 10% of disposable family income on out of pocket expenses. We identify individuals living in families that spend more than 5% or more than 10% of disposable family income on out of pocket expenses. Results are presented in terms of numbers or percent of individuals living in families with high financial burdens. Results are presented in terms of numbers or percent of individuals living in families with high financial burdens.

8. Results: Health Care Financial Burdens Among Medicaid Enrollees: 2004-05 14.6 million non-elderly persons in ‘Medicaid families’ 14.6 million non-elderly persons in ‘Medicaid families’ – subset of Medicaid population – ‘Medicaid family’ = all persons in the family were continuously enrolled in Medicaid or SCHIP 16.5% have high burdens 16.5% have high burdens – Spend 5% or more of income for health care 10.2% have very high burdens 10.2% have very high burdens – Spend 10% or more of income for health care

9. Comparison of Families Above/Below 5% Spending Threshold * * *P <.05 for difference between groups.

10. Components of OOP Spending In Families with High (5%) Burdens 1. Percent = (OOP spending for service / Total OOP spending) X 100

11. Contribution of Specific Services to the Risk of High (5%) Burden Sample = persons with a high (5%) burden Sample = persons with a high (5%) burden How many would continue to have a high burden if OOP spending for each service was set to zero? How many would continue to have a high burden if OOP spending for each service was set to zero?

12. Evaluating the Effects of State Cost Containment Policies We consider: prior authorization, generic substitution, copayments, quantity limits We consider: prior authorization, generic substitution, copayments, quantity limits Many states have multiple policies Many states have multiple policies Compare mean OOP RX spending in Compare mean OOP RX spending in – states with <3 polices – states with 3+ policies Use “raking post-stratification” weight adjustments to control for differences across policy groups Use “raking post-stratification” weight adjustments to control for differences across policy groups

13. Association of Cost Containment Policies with OOP Drug Spending * * *P <.05 for difference between groups.

14. Conclusion Many states have responded to financial pressures by implementing Medicaid pharmacy cost containment policies. Many states have responded to financial pressures by implementing Medicaid pharmacy cost containment policies. In implementing these polices, state programs may face a trade-off between In implementing these polices, state programs may face a trade-off between – reducing pharmacy costs – maintaining appropriate access to prescription drugs and shielding Medicaid enrollees from high spending burdens