Health Affairs, 24, no. 3 (2005): 780-789 doi: 10.1377/hlthaff.24.3.780 © 2005 by Project HOPE	New Online   Pay Cuts For Medicare Docs   Access To Care Woes   Public Coverage More Efficient   Empowering Consumers

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (14) Citing Articles via Google Scholar Google Scholar Articles by Cunningham, P. J. Search for Related Content PubMed PubMed Citation Articles by Cunningham, P. J. Related Collections Access To Care Consumer Issues Managed Care - Medicaid Medicaid Pharmaceuticals State/Local Issues Health Spending

Health Tracking

MARKETWATCH

Medicaid Cost Containment And Access To Prescription Drugs

Abstract

 
States have been intensifying their efforts to control rising prescription drug costs in their Medicaid programs. This study examines the effects of five Medicaid cost containment policies on enrollees’ perceptions of their ability to get prescription drugs. The results show that enrollees in states that have implemented all or almost all of these five policies have greater problems getting prescription drugs than enrollees in other states encounter. In terms of specific policies, prior authorization and mandatory generic substitutions had the largest effects on access to prescription drugs.

Despite these rising costs, all states have retained their outpatient prescription drug benefits, which is optional in Medicaid. To contain increasing costs, almost all states have implemented one or more strategies aimed at managing prescription drug use or curtailing wasteful and ineffective use. Cost sharing (that is, copayments) is one of the oldest and most prevalent forms of utilization management (UM), but a variety of other methods are increasingly being used. These include prior authorization, preferred drug lists, dispensing limits, mandatory use of generics, and step therapy.² Many states have also tried to contain costs by reducing their reimbursement to pharmacies, negotiating supplemental rebates with manufacturers, and monitoring high-cost users and prescribers of drugs.

As state budgets have tightened considerably since 2000—attributable in part to the increases in Medicaid spending as well as to slow economic growth—efforts to contain Medicaid drug spending have intensified. According to a recent survey, the percentage of states with mandatory prescribing of generics increased from 36 percent in 2000 to 70 percent in 2003.³ The percentage of states with dispensing limits increased from 20 percent in 2000 to 33 percent in 2003, while the use of cost sharing increased from 66 percent of states in 2000 to 81 percent in 2003. Moreover, the percentage of states using three or more cost-control strategies has increased greatly, from about one-third of states in 2000 to about 90 percent in 2003.

Previous research. Despite the near-universal implementation of one or more cost containment policies, virtually nothing is known about the effects of these policies on prescription drug costs, use, and access. Most of the prior research on this topic dates from the 1970s and 1980s and has focused on the effects of copayments on Medicaid prescription drug use and spending. The findings from these studies show that even a small copayment in Medicaid (for example, fifty cents per prescription) reduces prescription drug use and spending across a broad range of therapeutic categories, including those considered to be essential medications.⁴

Nevertheless, inferring the effects of recent cost containment policies on prescription drug use and access from this previous research is questionable, given the rapid increase in the development of new medications, use, and spending since many of these studies were conducted. And since most states have now implemented more than one cost containment policy, it is increasingly difficult to attribute effects on use and access to any specific policy. A more recent study (by this author) found that prescription drug access was particularly problematic in states that had implemented multiple cost containment policies.⁵

Present study. This study, using data from the 2000–01 and 2003 Community Tracking Study (CTS) household surveys, compares the extent of prescription drug access problems for adult Medicaid enrollees with the access problems of adults who have other types of insurance coverage and who have no insurance. It then examines the effects of cost containment policies and other factors on prescription drug access among adult Medicaid beneficiaries and the effect of the increase in states’ cost containment policies on changes in access between 2000–01 and 2003.

Study Data And Methods

Top Study Data And Methods Study Findings Discussion NOTES

 
Data. The CTS household survey is designed to produce representative estimates of health insurance coverage, access to care, use of services, and perceived quality of care for the U.S. population and sixty randomly selected communities in thirty-four states and the District of Columbia.⁶ The CTS is primarily a telephone survey, supplemented by in-person interviews of households without telephones, to ensure representation. Four rounds of the survey have been conducted since 1996; the most recent was completed in 2003. This study includes data from the third round of the survey (completed in 2000–01) and the fourth round (completed in 2003).

The overall sample for the surveys includes about 60,000 people in the 2000–01 survey and about 46,600 people in the 2003 survey. This study focuses on all adults (age eighteen and older) who reported that they were enrolled in Medicaid, which includes about 1,600 people represented in each of the two surveys.

Measure of drug access. Based on a similar question asked in the National Health Interview Survey (NHIS), sampled people in both the 2000–01 and 2003 surveys were asked the following question: "During the past twelve months, was there any time you needed prescription medicines but didn’t get them because you couldn’t afford it?" Although the question refers only to medications prescribed by a physician, the self-reported measure is limited in that the medical necessity of the medication could not be verified.

The strength of the measure is that it allows for broad comparisons of drug access between population groups. In this sense, validity is demonstrated by the fact that it is strongly correlated with characteristics of populations that are expected to have the greatest problems affording prescription drugs: namely, poor and low-income people, uninsured people, and people in poor health and with chronic conditions (that is, the most frequent users).

Measures of Medicaid cost containment policies. The CTS survey data are linked to data from state surveys of Medicaid prescription drug policies sponsored by the Henry J. Kaiser Family Foundation.⁷ The surveys were conducted in 2000 and 2003 and thus are roughly consistent with the time periods covered in the CTS household surveys. The surveys covered a number of policies related to utilization management and review, payment and purchasing policies, and institutional policies. This study focuses on five commonly used UM policies that are likely to have the most direct impact on enrollees’ use and access.⁸ These include the following: (1) Prior authorization: Pharmacists are required to obtain approval from the state (or a subcontractor) before dispensing a drug. (2) Copayments: Beneficiaries are required to pay a nominal copayment for each prescription (usually between fifty cents and three dollars), although the state cannot deny a drug because of inability to pay the copayment. (3) Dispensing limits: States limit the number of prescriptions that Medicaid will cover during a certain time period (for example, one month). (4) Generic drugs: State law requires generics.⁹ (5) Step therapy: Physicians need to demonstrate that a lower-cost drug is ineffective before prescribing a more costly alternative.

There is considerable variation among states in how these individual policies work (for example, states use different copayment amounts, and the medications subjected to prior authorization and generic mandates differ across states). Accounting for all of the variation in state policies is not feasible, and it is unlikely that the general measure of prescription drug access used in this study would be sensitive to many of these details.

Analysis. The study first compared access to prescription drugs for adult Medicaid beneficiaries with that of other adults (both insured and uninsured) in 2003, and then it examined changes in prescription drug access between 2000–01 and 2003. Estimates of access by insurance coverage were also computed by controlling for a variety of other factors that are likely to be correlated with access, including age, sex, race/ethnicity, family income, family type, health status, and chronic conditions.¹⁰

Regression analysis was used to examine the effects of Medicaid cost containment methods and other factors on access to prescription drugs for adult Medicaid beneficiaries.¹¹ As a general test of the intensity of states’ cost containment efforts, the analysis first examined the effect of multiple cost containment policies on access.¹² Observations from the 2000–01 and 2003 surveys were pooled into a single regression to increase the statistical precision of estimates. The analysis controlled for differences in individual characteristics and all other state-level factors (for example, other state policies, including state pharmacy assistance programs for dual eligibles).¹³

A second regression analysis was performed to examine the effects of each individual cost containment policy on access. This regression was identical to the first in all respects except that Medicaid cost containment policies were specified as five separate indicators for each policy. Examining and interpreting the effects of each policy is difficult because virtually all states have implemented more than one policy, and there is some inter-correlation in the effects of these policies on access.

Study Findings

Top Study Data And Methods Study Findings Discussion NOTES

 
Characteristics of Medicaid enrollees. Compared with all U.S. adults, Medicaid enrollees are somewhat younger, much poorer, more likely to be members of a racial/ethnic minority, and likely to have less favorable health overall and a much higher prevalence of chronic conditions (Exhibit 1). Lower incomes and high rates of health problems are particularly high-risk factors for problems with affording prescription drugs.

Problems getting prescription drugs. More than one-fifth of adult Medicaid enrollees reported that in 2003 they did not get prescription drugs because of cost (Exhibit 2). The extent of access problems for Medicaid enrollees was more similar to that of people who typically have no prescription drug coverage, including Medicare enrollees with no supplemental coverage and uninsured people. Those with private insurance, including Medicare enrollees with supplemental private insurance, have by far the lowest rate of drug access problems.

When income, health status, and other factors are controlled for (the adjusted estimates), rates of prescription drug access problems for Medicaid beneficiaries are much lower and more comparable with those of people with private insurance coverage (Exhibit 2). As one would expect, those who lack drug coverage entirely, including the uninsured and Medicare enrollees with no other coverage, have much higher levels of prescription drug access problems than those covered by Medicaid and private insurance.

There was no change between 2000–01 and 2003 in drug access for Medicaid enrollees (findings not shown). Access problems increased slightly for those with employer-sponsored private insurance (about one percentage point) and decreased for uninsured people (about three percentage points).

Changes in Medicaid cost containment policies. A large, growing percentage of Medicaid enrollees live in states that have implemented the various cost containment policies (Exhibit 3). Perhaps even more important is the dramatic increase in the percentage of Medicaid enrollees who live in states with four or five cost containment policies, from about 16 percent in 2000–01 to 62 percent in 2003. By contrast, the percentage of enrollees who live in states with two cost containment policies or fewer decreased from about 34 percent in 2000–01 to 7 percent in 2003.

Other factors affecting drug access. It seems probable that changes in other factors have offset the effects of increases in Medicaid cost containment policies, since there was no overall change in Medicaid prescription drug access between 2000–01 and 2003. Changes in key characteristics of Medicaid enrollees (noted in Exhibit 1) are one possibility, since several of these characteristics (such as age, family income, and number of chronic conditions) also have strong effects on drug access.

However, as Exhibit 4 shows, the net changes in access associated with these individual characteristics are negligible. For example, having multiple chronic conditions strongly increases the probability of having prescription drug access problems (16.3 percent). While a decrease in the percentage of Medicaid enrollees with multiple chronic conditions resulted in a net decrease in access problems, the effect was less than 1 percent. Other net effects of changes in individual characteristics were even smaller, and some changes (in age and income) led to a small net increase in access problems. When summed across all individual characteristics, there was no net effect on access from changes in enrollee characteristics.

On the other hand, the results show an unexplained decrease in the probability of prescription drug access problems between the 2000–01 and 2003 surveys (as indicated by the variable for the 2003 survey). In other words, some other factor(s) not directly accounted for in the regression analysis worked to reduce access problems, offsetting the effects of the increase in Medicaid cost containment policies.

It is unclear what these other factors could be, given the relatively short time period between the two surveys. One possibility is that even as states were becoming more aggressive in implementing cost containment policies, both health care providers and Medicaid enrollees became more knowledgeable and savvy about how these policies worked and therefore better able to navigate the system to get the prescription drugs they needed. It is also possible that states have gained greater experience in implementing policies and have refined them somewhat to reduce the access problems that some beneficiaries have had.

It may also not be coincidental that there was a small decrease in prescription drug access problems among the uninsured during the same time period. Improvements in safety-net capacity and coordination during this period may have benefited both Medicaid enrollees and uninsured people in providing greater access to free or lower-cost prescription drugs, as well as in providing more assistance to Medicaid enrollees in navigating the increasingly complicated rules of the Medicaid pharmacy benefit.¹⁴

Effects of individual cost containment policies. A second regression analysis—identical to the first in all respects except for the specification of Medicaid cost containment policies—was performed to test the effects of individual cost containment policies. Indicators for each of the five cost containment policies (whether or not the enrollee lives in a state with that policy) were included in the analysis. The results show that prior authorization and generic requirements had the largest negative effects on access: Prior authorization requirements increased the probability of prescription drug access problems by twenty percentage points; requiring the use of generic substitutes increased the probability by about eight percentage points (data not shown). Copayments, dispensing limits, or step therapy requirements had no statistically significant affects on access.

Prior authorization. The effects (or lack thereof) of individual policies are difficult to interpret, since most states have implemented more than one policy and there is some inter-correlation between these factors and the measure of prescription drug access. Since some form of prior authorization is used in most states, the large effects observed for this policy may be more indicative of the uniqueness of the small number of states that do not have such requirements. States without prior authorization tend to have very few cost containment policies of any type, and therefore prescription drug access problems would be expected to be much lower. Nevertheless, case studies of states’ prior authorization programs have observed that these programs can lead to bureaucratic and communication problems among enrollees, providers, and pharmaceutical benefit management firms under contract to the state, which in turn can lead to delays and other problems with prescription drug access.¹⁵

Generic mandates. That mandatory generic requirements would lead to higher reports of access problems may also be surprising, since these do not imply a cost to the enrollee. It is possible that some physicians continue to prescribe some brand-name drugs when generics are required because they are unaware of the policy or they believe that a brand name is more efficacious than its generic substitute. Also, some enrollees may specifically request brand-name drugs. If such drugs are prescribed when generic substitutes are required, enrollees would potentially have to pay the full cost of the medication.

Other policies. That copayments, dispensing limits, and step therapy did not have independent effects on prescription drug access may be because they are correlated with prior authorization and mandatory generic policies (that is, states often have both policies), or because these policies affect relatively few people and both the sample of Medicaid enrollees and the measure of access are too broad to observe the effects of these policies. For example, copayments and dispensing limits may affect only the very heaviest users of prescription drugs (of which there may be too few in the sample to affect the overall results), whereas step therapy applies to a relatively small number of drug classes.

Discussion

Top Study Data And Methods Study Findings Discussion NOTES

 
From the perspective of Medicaid enrollees, states’ efforts to contain the rising costs of prescription drugs are having negative effects on their access to prescription drugs. Although it is not possible with the general measure of access used in this study to distinguish between essential and nonessential medications, previous research on the effects of copayments and prior authorization have shown negative effects on use of prescription drugs across a wide range of therapeutic categories, including those considered "essential."¹⁶ That these access problems are particularly high for people with chronic conditions—which include a disproportionately high number of Medicaid enrollees—also suggests that there is unmet need for at least some essential medications.

Ongoing increases in Medicaid prescription drug costs and strained state budgets mean that cost containment policies will likely continue and even increase in states that have so far been less aggressive. Rolling back these policies is probably unrealistic given the magnitude of Medicaid spending on state budgets, especially when most private insurance plans are trying to contain costs through greater enrollee cost sharing, preferred drug lists, and incentives to use generics. Rather, policymakers should examine whether these cost containment policies are inadvertently restricting access to needed medications through bureaucratic obstacles or lack of awareness among Medicaid enrollees and providers as to how to navigate the often complex processes of obtaining approval to use certain drugs.

For example, pharmacists are not permitted by law to deny prescriptions to Medicaid enrollees even if they cannot pay the copayment, although research has shown that many pharmacists have poor knowledge of these requirements and often collect copayments when it is inappropriate to do so (for example, from children under age eighteen).¹⁷ Also, prior authorization programs in particular vary considerably across states, in terms of both the number and the types of medications that are subject to preauthorization as well as the processes that are established for obtaining authorization for a medication. Although this study suggests that prior authorization has resulted in reduced access, more intensive case studies of state programs reveal considerable variation in how they operate as well as the level of difficulty enrollees and their physicians have in obtaining approval.¹⁸

Identifying the exact mechanisms by which the various cost containment policies affect access was not possible in this analysis because of the relatively small samples, the general measure of access, and difficulty in developing more detailed classifications of state policies based on available data. Further evaluation and monitoring of the effects of cost containment policies on access at the state level are key to successful implementation and streamlining of these policies.

Ensuring that cost containment policies do not inadvertently harm access to necessary medications is especially important given the high need for prescription medications among Medicaid enrollees as well as their lower ability to use their own resources to make up for any shortfall.

Editor's Notes

 
Peter Cunningham (pcunningham{at}hschange.org) is a senior health researcher at the Center for Studying Health System Change in Washington, D.C.

This research was supported by the Robert Wood Johnson Foundation through its support of the Center for Studying Health System Change. The author thanks Paul Ginsburg, Len Nichols, and Jim Reschovsky for providing helpful comments, and Ellen Singer of Social and Scientific Systems for excellent programming assistance.

NOTES

Top Study Data And Methods Study Findings Discussion NOTES

 

1. B. Bruen and A. Ghosh, Medicaid Prescription Drug Spending and Use (Washington: Kaiser Commission on Medicaid and the Uninsured, June 2004).
2. J.S. Crowley, D. Ashner, and L. Elam, Medicaid Outpatient Prescription Drug Benefits: Findings from a National Survey, 2003 (Washington: Kaiser Commission, 2003).
3. Ibid.
4. C.E. Reeder and A.A. Nelson, "The Differential Impact of Copayment on Drug Use in a Medicaid Population," Inquiry 22, no. 4 (1985): 396–403[ISI][Medline]; B. Stuart and C. Zacker, "Who Bears the Burden of Medicaid Drug Copayment Policies?" Health Affairs 18, no. 2 (1999): 201–212[CrossRef][Medline]; S. Soumerai et al., "Payment Restrictions for Prescription Drugs under Medicaid: Effects on Therapy, Cost, and Equity," New England Journal of Medicine 317, no. 9 (1987): 550–556[Abstract]; and A.A. Nelson, C.E. Reeder, and W.M. Dickson, "The Effect of a Medicaid Drug Copayment Program on the Utilization and Cost of Prescription Services," Medical Care 22, no. 8 (1984): 724–736.[ISI][Medline]
5. P.J. Cunningham, Affording Prescription Drugs: Not Just a Problem for the Elderly (Washington: Center for Studying Health System Change, April 2002).
6. R. Strouse, B. Carlson, and J. Hall, Community Tracking Study: Household Survey Methodology Report 2000–01 (Round Three), Technical Publication no. 46, July 2003, www.hschange.com/CONTENT/602/602.pdf (16 February 2005).
7. Crowley et al., Medicaid Outpatient Prescription Drug Benefits; and R. Schwalberg et al., Medicaid Outpatient Prescription Drug Benefits: Findings from a National Survey and Selected Case Study Highlights (Washington: Kaiser Commission, October 2001).
8. A notable exclusion from these five policies is the use of preferred drug lists (PDLs), which are excluded because they were ascertained in the 2003 Kaiser Family Foundation survey but not the 2000 survey, whereas the five policies included in the analysis were ascertained in both surveys. PDLs are in many ways a variant of prior authorization (prior authorization is needed for drugs not on the list), and therefore the exclusion is unlikely to have any major effects on results. In fact, separate analyses of 2003 data only (based on the methodology described below) did not find any independent effect of PDLs on prescription drug access.
9. In some states, the use of generics is encouraged (but not required) through the use of differential copayments, differential dispensing fees, and provider education. The measure in this study reflects only states that require generics as the most restrictive form of cost containment policies.
10. Adjusted means were computed from ordinary least squares (OLS) regression using the 2003 survey data, with access as the dependent variable and insurance coverage and other individual characteristics as independent variables.
11. Although a probit or logistic regression is normally used when the dependent variable is binary, OLS regression was used in this analysis because the coefficients reflected linear probabilities and were therefore easier to interpret. Results from a logistic regression analysis produced very similar results.
12. The analysis focused on comparing states that have implemented four or five policies with states that implemented three or fewer. There were too few states that had implemented no policies to analyze separately, and earlier versions of the analysis show that there were no differences in access between states with two or fewer policies and states with three policies.
13. Dual eligibles are people who are eligible for both Medicaid and Medicare. Binary variables for each state were included to control for all other state-level factors that could potentially confound the effects of cost containment policies on prescription drug access. A similar indicator for survey year was included to account for all other secular changes in access not accounted for by the variables in the model.
14. J.F. Hoadley, L.E. Felland, and A.B. Staiti, Federal Aid Strengthens Health Care Safety Net: The Strong Get Stronger, Issue Brief no. 80 (Washington: HSC, April 2004).
15. J. Tilly and L. Elam, Prior Authorization for Medicaid Prescription Drugs in Five States: Lessons for Policy Makers (Washington: Kaiser Commission, April 2003).
16. Soumerai et al., "Payment Restrictions"; and Reeder and Nelson, "The Differential Impact of Copayment."
17. C. Fahlman, B. Stuart, and C. Zacker, "Community Pharmacist Knowledge and Behavior in Collecting Drug Copayments from Medicaid Recipients," American Journal of Health System Pharmacy 58, no. 5 (2001): 389–395.[Abstract/Free Full Text]
18. Tilly and Elam, Prior Authorization.

                      What's this?

This article has been cited by other articles:

				T. J. Johnson and S. Stahl-Moncada Medicaid Prescription Formulary Restrictions and Arthritis Treatment Costs Am J Public Health, July 1, 2008; 98(7): 1300 - 1305. [Abstract] [Full Text] [PDF]

				D. P. Goldman, G. F. Joyce, and Y. Zheng Prescription Drug Cost Sharing: Associations With Medication and Medical Utilization and Spending and Health JAMA, July 4, 2007; 298(1): 61 - 69. [Abstract] [Full Text] [PDF]

				J. M. Polinski, P. S. Wang, and M. A. Fischer Medicaid's Prior Authorization Program And Access To Atypical Antipsychotic Medications Health Aff., May 1, 2007; 26(3): 750 - 760. [Abstract] [Full Text] [PDF]

				C. Smith, C. Cowan, S. Heffler, A. Catlin, and the National Health Accounts Team National Health Spending In 2004: Recent Slowdown Led By Prescription Drug Spending Health Aff., January 1, 2006; 25(1): 186 - 196. [Abstract] [Full Text] [PDF]