UCSF

BACKGROUND: Although medically necessary in some cases, there is growing concern that compounded medications are being overprescribed, leading to questions about safety and necessity for high use and cost. Safety concerns regarding compounded medications were highlighted by the 2013 contamination of steroid injections by the New England Compounding Center, which caused serious infections and other injuries to at least 751 patients and resulted in at least 64 patient deaths. A study contributed to our understanding of compounded medication use and cost, finding in a sample of commercially insured population that the average ingredient cost for compounded medication prescriptions was $710.36, which is 130% higher than for noncompounded medication prescriptions. The literature on use and cost of compounded medications in noncommercially insured populations and related regulations, however, is sparse. The California Workers Compensation System (CAWCS)-the largest U.S. workers compensation system and a public health system experiencing high compounded medication costs-provided an opportunity for additional analysis of these issues. Furthermore, CAWCS data on compounded medication use and cost allow for the exploration of alternative pricing mechanisms that may control costs. OBJECTIVES: To (a) examine use, cost, and billing and reimbursement practices for compounded medications in a public health system-CAWCS- and (b) evaluate regulations and recommend an alternative pricing mechanism that could control costs in California. METHODS: Descriptive statistics for use, cost, and reimbursement patterns of all compounded medication prescriptions included in CAWCSs Workers Compensation Information System claims datasets from 2011 to 2013 were determined. This study coded a unique dataset that (a) identified compounded medications at the ingredient level; (b) grouped compounded medications from ingredient level to compounded medications as a whole; and (c) categorized compounded medications into applicable Colorado pricing categories. T-tests assessed if regulation AB 378, which targets compounded medications, was associated with a difference in mean cost. The Colorado pricing scheme was applied to estimate cost and provide recommendations. RESULTS: Despite the AB 378 requirement for compounded medications to be billed at the ingredient level for reimbursement, 15% of pharmacy-dispensed and 6% of physician-dispensed medications were not billed at the ingredient level. For pharmacy-dispensed compounded medications billed at the ingredient level, mean amount paid (SD) per ingredient was $45.40 (195.97), and for those medications billed at the single compounded medication level, mean amount paid (SD) per medication was $95.20 (326.33) over all years. For physician-dispensed medications billed at the ingredient level, mean amount paid (SD) per ingredient was $75.47 (205.51), and when billed at the single medication level was $204.83 (221.01). T-tests showed a mean increase in compounded medication mean amount paid between pre- and post-AB 378 groups of $12.27 (P < 0.001) for pharmacy-dispensed medications and $11.34 (P < 0.001) for physician-dispensed medications, suggesting that AB 378 did not curb compounded medication mean amount paid. CONCLUSIONS: The average cost of CAWCS pharmacy- and physician-dispensed compounded medications consistently increased. Various factors may have influenced this increase, but AB 378 did not achieve its full regulatory intent to standardize billing and reimbursement and control cost. Grouping of ingredients into compounded medications allowed for application of the Colorado pricing scheme to CAWCS claims data. Adoption of Colorado pricing would save 46% of current compounded medication cost for less complicated medications, while increasing cost for more complicated medications. The analyses recommended a revised Colorado pricing scheme, which would provide improved incentives for accurate billing and lead to savings for CAWCS. DISCLOSURES: Funding for this study was provided by the California Workers Compensation System. The authors had final control regarding study design, study conduct, and writing of the manuscript. The authors report no conflicts of interest.