THE IMPACT OF SCHEDULING CONTROLLED SUBSTANCES
The Federal government has implemented policies to regulate drugs with abuse potential for over 100 years. The key law governing most controlled substance activities is the 1970 Comprehensive Drug Abuse Prevention and Control Act. This law created a scheduling system that impacts pharmaceutical drug development, manufacturing, prescribing and dispensing of medicines. Since its implementation over 40 years ago, there has been limited research into the value or effectiveness of the scheduling system.
The scheduling system includes placing controls that affect the healthcare system including the manufacture of medicines, the need for specialized nonclinical and clinical studies and limits on prescription refills. Despite the system existing for several decades, there is a paucity of data determining the how effective scheduling is and whether it meets the intended goal of limiting abuse and thereby preventing harm to the public. The purpose of this dissertation is to explore the impact of drug scheduling. The objectives of the three manuscripts of this dissertation were to: a) review the present scheduling system and understand its impact in drug development; b) examine the impact of scheduling on prescribers; c) determine the effectiveness of scheduling in reducing harm.
The first manuscript is a literature review conducted using Publisher Medline (PubMed) and ScienceDirect for scholarly articles regarding controlled substances and the scheduling system. Additionally, the accessible sites for the Department of Health and Human Services (HHS), the Food and Drug Administration (FDA) and the Drug Enforcement Agency (DEA) were used to perform a search of all relevant laws and regulations dealing with drug scheduling and control. Due to the limited number of scholarly articles on the scheduling system, the manuscript focused as an overview of the current scheduling system.
The second manuscript is a cross sectional survey designed to assess physician knowledge of the scheduling system and the reported level of impact in their prescribing of controlled substances. A random sample of 400 prescribers in two different sections of the country (Midwest and New England) was surveyed. The questionnaire had two parts – Part I included demographic information and Part II included questions addressing three areas (physician attitude, physician knowledge, prescribing impact). Statistical analysis of the responses included screening covariate variables to determine the importance factors. For each identified important factor (Chi-square test or linear regression model) was performed to measure the relevance and the P-value.
The third manuscript is a retrospective analysis study which utilized data from the Drug Awareness Warning Network (DAWN) and IMS Health to determine the ‘rate of harm’, defined as emergency department admissions, of established scheduled and nonscheduled drugs as well as a recently approved drug. These rates were then compared to a drug that was scheduled at a level between the mature drugs to determine if rate of harm was commensurate with schedule. A non-inferiority design was used to test whether three scheduled drugs were equivalent or superior (significantly higher rate of harm) to the nonscheduled drugs.
In the survey study, the majority of prescribers responded in a negative manner regarding scheduling for all survey questions. For physician attitude and knowledge, this ranged from disagreeing with the regulations not allowing refills for Schedule II (C-II) drugs to not understanding that C-V drugs do not have Federal refill restrictions. A majority of physicians also noted the controlled substance numbering system did not convey the information they needed for prescribing and that scheduling, in general, has a negative impact on patient care.
For the second study, five CNS-active drugs were tested in a non-inferiority design to determine rates of harm and comparing scheduled versus nonscheduled drugs. Four of the five drugs were established medicines with generic versions on the market and greater than a decade of availability from first launch. A baseline acceptable rate of harm range was set using the nonscheduled drugs, gabapentin and phenytoin. In all three years, 2006-2008, pregabalin, a C-V drug demonstrated an equivalent or lower rate of harm than the nonscheduled drugs when using data from the DAWN drug misuse and abuse category. The C-IV drugs, alprazolam and triazolam, did not show replicate evidence of equivalence to the reference drugs and exceeded the non-inferiority margins in most or all years.
In reviewing the history of controlled substances regulation, the genesis of most government action appeared to come from political or financial aspects rather than science or medicine. In the absence of seminal events such as major global treaties, which forced the creation the 1970 Controlled Substance Act, little time has been spent updating or monitoring the system Congress put into place. Further, there is a lack of scholarship in academic or governmental environments to determine the value, impact or accuracy of this system. This lack of concern regarding value or effectiveness of the system currently in place can be seen in the results of the studies performed for this dissertation.
The findings from the survey study demonstrate the lack of value for current system has for prescribers. The majority of prescribers found the system to be confusing and creating a negative impact on their delivery of healthcare. The second study identified a potential new tool for monitoring scheduled drugs. The non-inferiority test accurately identified the two C-IV drugs as associated with more harm (drug misuse/abuse emergency department visits) than either the C-V drug or the nonscheduled drugs. However, with limited data, the C-V drug (pregabalin) appeared to have at least equivalent and potentially a lower level of harm associated with it compared to the nonscheduled drugs. This could mean the scheduling of pregabalin is unjustified thus wasting already constrained medical resources. Using non-inferiority testing, the government could monitor marketed drugs to determine if, due to changes in the detected rate of harm, a drug needs a different (higher or lower) level of control assigned to it. Further research in this area is warranted.