Doctors who care for the poor: Paying the hidden cost of Medicaid

A groundbreaking study has finally put a dollar figure on a previously unanswered question: how much do physicians' practices, due to government regulation, pay to ensure their poorest patients get the right prescription drugs? Turns out the answer is $8.02 per prescription, on average, or $1,110 annually for high blood pressure and high cholesterol medications alone, according to research from Jonathan D. Ketcham, an assistant professor at the W. P. Carey School of Business, and Andrew J. Epstein, an assistant professor at the Yale School of Medicine.

A groundbreaking study has finally put a dollar figure on a previously unanswered question: how much do physicians' practices, due to government regulation, pay to ensure their poorest patients get the right prescription drugs? Turns out the answer is $8.02 per prescription, on average, or $1,110 annually for high blood pressure and high cholesterol medications alone, according to "Medicaid Preferred Drug Lists' Costs to Physicians," recently published in Medical Care, a health policy journal.

And in a second study looking at the same data, "Which Physicians Are Affected Most by Medicaid Preferred Drug Lists for Statins and Anti-hypertensives," researchers conclude that solo practitioners working in low-income communities with lots of Medicaid patients end up paying the most.

Hidden cost of regulation

Why and how are some physicians funding Medicaid, a taxpayer-funded, government-program benefit? The answer is complicated. In most states, when writing a prescription for a Medicaid patient, physicians must pick from a preferred drug list, or PDL, that relies heavily on generics and excludes many expensive and new medications.

If a doctor is convinced that a particular patient needs a non-PDL drug, he or she may ask permission to prescribe it anyway. The process of asking racks up office costs — an average of $12.22 per request — that are not reimbursed. These are in addition to un-reimbursed costs physician offices incur to keep up-to-date on their states' PDLs.

Results of the studies, conducted by Jonathan Ketcham, an assistant professor at the W. P. Carey School of Business, and Andrew J. Epstein, an assistant professor at the Yale School of Medicine, have far-reaching health policy implications. "The much broader story, beyond the strong altruistic motivation of these physicians, is the hidden cost of government regulation," Ketcham explained.

Policy-makers should weigh more carefully the benefits of regulation — like limiting the over-use of prescription drugs — against the off-the-books but very real burden of regulation, he and Epstein contend. Study results also carry important implications for future Medicare costs, because under Part D, Medicare supplanted Medicaid as the single biggest payer for prescription drugs. "The question is, how large will the physicians' administrative burden become if Medicare's new prescription drug program goes the way of Medicaid's?" Ketcham asked.

Physicians surveyed

He and Epstein surveyed cardiologists and primary-care physicians in nine states, asking about their Medicaid PDL prescribing habits for two classes of drugs — statins and anti-hypertensives. The researchers combined these survey results with pharmacy claims data on actual prescriptions filled.

Based on the first study's $8.02-per-prescription figure, they concluded that physicians in those 9 states and two specialties alone incurred slightly more than $53 million in PDL-related, un-reimbursable administrative costs in 2005 just for those two drug classes.

Now imagine that these estimates are extrapolated to include all classes of drugs — not just statins and anti-hypertensives — prescribed by doctors to all Medicare patients nationwide. What if Medicare's Part D prescription benefit is administered similarly? Ketcham and Epstein ran the numbers; the study notes that "similar restrictions under Medicare Part D across all therapeutic classes would have cost physicians $3.18 billion in 2006."

Supply-side controls

Mandating a PDL isn't a unique strategy; in fact, it's a common cost-cutting move shared by insurance companies and other healthcare payers. As Ketcham said, when patients can't be made to pay more for their care through co-pays and deductibles, "it limits the program's ability to control costs. The alternative is to institute supply-side controls, often through a preferred drug list."

It's also no secret that the controls generate large administrative costs for the physicians, he noted. "Physicians want to avoid incurring these additional administrative costs and therefore are more likely to prescribe from the PDL. It's an intentional motivator. What hasn't been quantified until this study is the actual amount, in dollars, that doctors are paying, which otherwise remains as part of the hidden cost of the Medicaid system," he continued.

To ensure that sick people get the necessary care, Medicaid decision-makers built in an exception policy called "prior authorization" that allows the doctor to go off the PDL on a case-by-case basis. If a prior authorization, or PA, is denied, the physician can appeal the decision.

Applying for or appealing a PA means the doctor and his staff spend extra time filling out forms, consulting with pharmacists and talking with patients. Updating the PDL and taking PDL training generates more un-reimbursable staff time. Some Medicaid PDLs require doctors to try a second-choice drug before allowing them to go off the list, a process also known as "fail-first" or "step-wise" therapy.

Again, this government cost-cutting strategy hits physicians in the pocket, incurring staff time that is un-reimbursable. Ketcham and Epstein note that in a few cases — roughly 4 percent of the time — frustrated physicians eventually decide against prescribing any drug at all. When PDL-related tasks were ranked by administrative cost, "requesting a PA" came in first.

And doctors request PAs often; in fact, 24 percent of the new prescriptions written by the surveyed doctors were attached to PAs. Next was "communicating with pharmacists and patients," followed by "appealing PA rejections," "preference costs due to prescribing no drug or a second-choice drug," tracking the PDL and receiving training.

The hardest hit

The first study found that the hidden cost of Medicaid's PDL program averaged $1,110 per physician for the two large drug classes alone. But actual costs varied from state to state, depending on factors such as volume of Medicaid prescriptions written, the percentage of drugs not covered by PDLs and other characteristics of the individual state's PDL.

Physicians in Massachusetts incurred costs averaging of $635 annually, compared to South Carolina doctors, who incurred $1,596. It seems likely costs per physician could decrease if higher-cost states adopted the PDL characteristics of lower-cost states, the researchers suggest. In the second study, which was published in late 2006, Ketcham and Epstein analyzed which doctors were hit hardest when prescribing statins and anti-hypertensive drugs to Medicaid patients in a PDL environment.

They found that physicians practicing alone, rather than in a group, paid out more when dealing with PDLs. Solo doctors who worked in low-income communities and had a high percentage of Medicaid patients fared the worst. Although the PDL rules affected how all the physicians prescribe meds, it affects the prescribing habits of doctors with the highest volume of Medicaid prescriptions the most.

One notable exception was that physicians in the least disadvantaged areas were the most likely to close their doors to Medicaid patients due to the burden of PDLs. The authors were careful to note that generalizing their study might not be possible due to their choices of the studies' states, specialties, and types of medications. They also tested the validity of the data that physicians self-reported by comparing it with other, external sources.

These comparisons indicated that the costs they estimated were consistently lower than those implied by other data sources. "What we're saying is that the hidden costs to physicians need to be part of the greater conversation. Clearly, adding these quantified costs can help Medicaid and Medicare administrators make more informed decisions.

Without this information, I suspect they would have continued to ignore the costs to physicians," Ketcham concluded. Epstein further emphasized that, "physicians who work predominantly with poor Medicaid patients are hit the hardest by these regulations, which is not surprising but seems inherently unfair. It also raises longer-term concerns about poor patients' access to both prescription drugs and physician services."

Bottom Line:

• State Medicaid programs' use Preferred Drug Lists to control spending on prescription drugs. This generates substantial administrative costs for physicians. These costs fall disproportionately on those who see more poor, minority patients.
• The administrative costs average $8.02 per prescription written. If Medicare Part D implemented a Preferred Drug List nationwide, it would have cost physicians an estimated $3.2 billion in 2006.
• Preferred drug lists also influenced physicians' prescribing patterns and reduced physicians' willingness to treat Medicaid patients.
• Government regulations can create large costs for businesses including health care providers, and these costs must be included in the estimates of government policies' total costs.