Is Prior Authorization for Prescribed Drugs Cost-Effective?

With the increasing managed care restrictions on health care coverage, a look into the efficacy of some of these procedures is needed. This study examines the cost burden of implementing prior authorization (PA) for prescription drugs. Seventy-five prescriptions dispensed by 2 Philadelphia pharmacies requiring PA were tracked and the savings analyzed. Requiring PA proved to be financially beneficial to MCOs but resulted in an increase in uncompensated time for physicians and pharmacies. Two classes of drugs, antihistamines and proton pump inhibitors, accounted for 48% of the medications requiring PA. By educating plan members in advance concerning coverage limits for these medications, MCOs could decrease the use of PA and reduce the amount of time spent by physicians, pharmacists, and patients in dealing with this procedure. (Drug Benefit Trends. 2008;20:136-139)

MCOs are restricting coverage to contain rising health care costs. Methods used include implementing or increasing patient copayments or coinsurance, restricting use of out-of-network care providers and choice of hospital, requiring prior approval for some diagnostic and surgical procedures, and eliminating coverage for experimental treatments. For the pharmacy benefit, cost containment methods include implementing or increasing patient copayments or coinsurance, formulary restrictions, setting quantity limits, incentivizing generic drug use, and implementing prior authorization (PA), the focus of this report.

PA requires that physicians complete a form and submit it to the MCO for approval before prescribing the medication. PA may be implemented to limit use of new, costly, or potentially toxic medications and substitute less costly or safer options.¹ PA is also used to ensure appropriate clinical use of medications such as human growth hormone. In addition, PA may be used for new medications until an MCO can conduct a thorough evaluation and compare the new agent with medications currently on the market. However, complying with the PA process puts a burden on the physician and the pharmacy. PA limits the range of medications physicians can prescribe, takes valuable time, and delays patient treatment. PA is similar to the preauthorization process implemented by MCOs that physicians must follow to justify ordering costly tests or surgical procedures. To obtain preauthorization, physicians must first phone a case manager at the MCO, often a nurse practitioner, and explain what treatments have already been attempted unsuccessfully and why the proposed treatment or diagnostic test is needed.

For pharmacists, PA takes time that could be spent filling other prescriptions and providing more patient counseling. In one study, a supermarket pharmacy spent 2.25 minutes and an independent pharmacy spent 2.97 minutes on average (uncompensated) processing each rejected prescription that required PA.¹

In another study, dispensing prescriptions that required pharmacist intervention to resolve reimbursement-related problems (such as nonformulary drug, plan limits exceeded, override needed for early refill, or PA) required 3.4 times more time spent by the pharmacist compared with the time needed to dispense a prescription without any coverage issues.² The researchers concluded that the additional time required to dispense prescriptions with coverage issues would justify a dispensing fee of $12 per prescription instead of the actual dispensing fee paid of $3.50 per prescription.² Because MCOs do not compensate care providers for following the PA process, pharmacies would incur savings if PA was eliminated.

The PA process, which can take from a few hours to a few days to complete, can also cause delays for patients in receiving needed medications. PA can decrease patients' satisfaction with their pharmacy, which is based in part on short wait times and being able to pick up the prescription at the promised time. The PA process requires patients to wait longer for the prescription to be filled and sometimes causes confusion when the patient cannot receive the medication prescribed. Patients may blame the pharmacist for refusing to dispense the medication that their doctor prescribed when in fact such decisions are made by the MCO.

With the increased burden that the PA process puts on health care providers, some question its effectiveness in reducing costs. With most prescriptions subject to PA ending up being approved and the high administrative costs of processing these claims, how cost-effective is this process? In one study, researchers found that more than 95% of drugs requiring PA were approved for the original drug and only 4.4% were ultimately denied.³

Some MCOs have recently discontinued requiring PA for select medications, including Viagra, Zyban, and cyclooxygenase-2 inhibitors, after finding that the administrative costs offset any savings resulting from fewer prescriptions being dispensed.⁴ Given these findings, it is not clear that the cost-effectiveness of the PA process for prescription drugs has been clearly established-even for MCOs.

Methodology

A pilot study was undertaken to test the hypothesis that the PA process is not financially viable and is detrimental to patient care. To test this hypothesis, 75 prescriptions requiring PA dispensed by 2 community pharmacies in the Philadelphia area were identified during a 2-month period in 2007. These prescriptions were later checked to see if the PA was approved or denied and whether the patient's medication was changed. If the therapy was changed to another drug, any savings for the MCO was noted. For prescriptions that were denied, it was noted whether MCO coverage extended to an alternative product.

Results

Of the 75 prescriptions requiring PA, 52 (69.3%) were approved and 23 were denied. This proportion of 69.3% did not approach the over 95% approval rates that some other researchers have found.³ Of the 23 PAs that were denied, 6 prescriptions were changed to other appropriate therapy covered by the MCO, while no alternative coverage was provided for the other 17.

By switching patients to alternative medications in these 6 cases, the MCOs saved $399.20 or an average of $66.50 per prescription. For the 17 prescriptions where the patient either paid out-of-pocket for an alternative or went without medication, PA resulted in savings for the MCO of $1445 (17 times $85), taking into account that a prescription filled in 2007 averaged $85. Total savings for the MCO was $1844.20 ($399.20 for the 6 prescriptions that were switched to another therapy and $1445 for the 17 denied prescriptions at an average cost of $85 each). Thus, the amount saved by the MCO would clearly cover the administrative costs in these 75 cases. However, the denial of coverage for alternative medications may turn out to be detrimental to the patient's health.

The burden of the PA process was found to be concentrated on certain classes of drugs (Table). Nearly half (48%) of PA requests were for allergy medications (26.67%) or proton pump inhibitors (21.33%). The drug class for which MCOs most frequently denied PA requests and declined to cover a substitute medication was allergy medications, including Zyrtec and Allegra. Zyrtec required PA 9 times. Of these 9, the PA request was approved 5 times. However, for the other 4 times the medication was denied, the MCO declined to pay for any alternative therapy. Five PAs for Allegra prescriptions and 2 for Allegra D were approved. Two PAs for Allegra D were denied for which the MCO offered no alternative coverage. Thus, on 7 occasions, patients of these MCOs received no coverage for allergy medications.

MCOs also frequently decline coverage of prescriptions for proton pump inhibitors. PA was required for 10 Nexium prescriptions. On 3 occasions, PAs for Nexium were approved and twice the prescription was changed to generic omeprazole. The other 5 PAs resulted in the patients not receiving any medication covered by the MCO. Prescriptions for the proton pump inhibitor AcipHex required PA twice and, in both instances, the prescription was denied without any covered alternative offered. One prescription for omeprazole required PA and it was denied with no alternative therapy coverage offered. On a total of 8 occasions, patients were denied treatment with proton pump inhibitors through the PA process and they received no alternative covered therapy.

Discussion

From this research, it is clear that implementation of PA results in cost savings for MCOs. By denying patients coverage of medications for which OTC alternatives are available, MCOs save money by not paying for these medications. However, for patients who do not receive medication for their conditions, more serious problems could potentially develop later on, which could result in higher overall health care costs for MCOs. Although allergy medications and proton pump inhibitors have OTC alternatives, patients may not be familiar with these medications or may not want to purchase them. In addition, the PA process is time-consuming for physicians and community pharmacies and inconvenient for patients.

The burden on physicians, pharmacies, and patients related to complying with PA could be reduced if MCOs clearly explained their coverage policy for those therapeutic categories most affected. By describing to members how certain medications will not be covered because OTC alternatives are available or the circumstances under which they will be covered, the company could avoid patient complaints about drugs not being covered and reduce the time physicians and pharmacies spend to satisfy PA requirements. MCOs would also benefit in that they could lower their administrative costs by eliminating the need to handle so many PA requests. Thus, by clearly explaining the plan's policies on these commonly prescribed medications that require PA, MCOs could reduce their administrative costs while easing the burden placed on physicians and pharmacies to comply with the PA process, and it could also diminish the frustrations of customers who may be left inconvenienced or even confused by the process.

Limitations

There are several limitations to this pilot study. The study was conducted at only 2 pharmacies in Philadelphia, which limits the extent to which findings are generalizable. Only 75 prescriptions requiring PA were studied over a 2-month period. There was also no randomization in this study. Despite these limitations, some important findings can be extracted.

Conclusion

PA saves money for MCOs despite the administrative costs that are incurred by the process. However, costs are increased for physicians and pharmacies that implement PA on behalf of MCOs. The PA process in this study was found to be primarily focused on the use of antihistamines and proton pump inhibitors, for which there are OTC alternatives that are generally not covered by the MCO. Through better communication of coverage policies to members, MCOs can ease the burden of the PA process for care providers while also reducing MCO administrative costs and patients' frustrations with the process.

References:

References

• 1. MacKinnon NJ, Kumar R. Prior authorization programs: a critical review of the literature. J Manag Care Pharm. 2001;7:297-303.
• 2. Wertheimer AI, Santella TM. Effect of prior authorization and formulary limitation on community pharmacy practice and profitability. Drug Benefit Trends. 2006;18:36-49.
• 3. LaPensee KT. Analysis of a prescription drug prior authorization program in a Medicaid health maintenance organization. J Manag Care Pharm. 2003;9:36-44.
• 4. Titlow K, Randel L, Clancy CM, Emanuel EJ. Drug coverage decisions: the root of dollars and values. Health Aff. 2000;19:240-247.