|
BACKGROUND/RATIONALE:
BACKGROUND: Prescription opioid abuse is a growing and high profile problem in the United States, and is associated with substantial loss of physical and mental health functioning and risk of overdose death. According to SAMHSA’s National Survey on Drug Use and Health in 2004, 31,768,000 Americans over the age of 12 had abused a prescription opioid at least once in their lives, with 4, 404,000 having done so in the past month. Injuries in returning OIF/OEF veterans and aging in the existing veteran population will increase legitimate need for opioid medication for chronic pain, thus increasing veterans’ exposure and access to these medications, and raising the risk of abuse. Prescription opioid abusers often manipulate clinicians and take advantage of poor care coordination and information systems to obtain prescription opioid medication. The signs of prescription medication abuse include characteristic patterns of health care utilization and diagnostic histories, such as high use of the emergency room and a history of substance use disorder. With its computerized medical record system, the VA is particularly well-equipped to identify these patterns of health care utilization, and could use this information to warn clinicians of potentially aberrant behavior. HSR&D Grant TRX 04-402 “Decision Support for the Management of Opioid Therapy in Chronic Pain” has been developing a computerized decision support system, based upon the VA/DoD Clinical Practice Guideline (CPG) for the Management of Opioid Therapy for Chronic Pain, to assist primary care clinicians with opioid prescribing. One objective of this system is to warn clinicians when patients are at risk of or display signs of prescription opioid abuse. Based upon the CPG and existing medical record variables, we are developing algorithms for flagging high-risk patients for further assessment. However, the literature is generally based upon expert opinion rather than validated methodology. To develop a functional algorithm, general recommendations must be operationalized: for example, how many ER visits over what frequency is suggestive of abuse? To reliably distinguish between high- and low-risk patients, medical record-based assessment algorithms must be validated on a sample of patients receiving opioids.
OBJECTIVE(S):
OBJECTIVE: The primary objective of this project is to develop and validate an algorithm to identify patients at risk of or with signs of prescription opioid abuse based up on information in the VA computerized medical record. This algorithm will be immediately implemented at the VA Palo Alto as part of a pilot study of a computerized decision support system (ATHENA-Chronic Pain) to assist primary care clinicians with opioid prescribing.
METHODS:
METHODS: Here we propose to conduct interview assessments of aberrant prescription opioid use behaviors in a population of 200 VA Palo Alto patients who received at least 2 prescriptions of opioid medication for non-cancer pain in the last year. Aberrant opioid medication use scores based upon these assessments will be compared with results of the medical record-based algorithm and the algorithm will be optimized to best predict aberrant behaviors. If a strictly medical record-based algorithm that adequately predicts aberrant opioid medication use behaviors cannot be created, we will examine whether will examine whether addition of additional screening questions suggested as having utility based on the responses to the evaluations could substantially improve detection of opioid misuse.
FINDINGS/RESULTS:
No results at this time.
IMPACT:
IMPACT: The optimized algorithm will be incorporated into ATHENA-chronic pain, disseminated throughout VA as an internal report, published in the scientific literature, relevant computer code will be made available, and future projects will evaluate the clinical impact of this warning system when implemented within VA HCSs. Early identification and targeted monitoring of patients with signs of misuse may improve clinical care for at risk patients and reduce medical and legal problems associated with abuse.
PUBLICATIONS:
None at this time.
DRA:
Substance Abuse, Addictive Disorders
DRE:
Prevention
Keywords:
Drug abuse, Risk factors, Informatics
MeSH Terms:
none
|
