Chronic Pain Agreement Violations in Patients with Cancer; Incidence and Associated Risk Factors
Justin Tokorcheck MD, Brandon Seifert MD, Kristopher Atwood PhD, Oscar DeLeon-Casasola MD
SUNY University at Buffalo, Roswell Park Cancer Institute
Introduction and Objectives
While at present there are multiple strategies, modalities, and medications available for the treatment of chronic pain, patients who carry a cancer diagnosis are often treated with a combination of approaches to alleviate symptoms. Among those who suffer from cancer pain, opioids are the mainstay for treatment1. While opioid treatments can be effective, the institution of such therapeutic plans requires careful consideration and assessment of individual compliance with the prescribed regimen in light of the opioid abuse epidemic2. Noncompliance or abuse of opioid therapies can conflict with stringent state and federal laws in place governing the use of controlled substances and also carries the potential for serious adverse health effects. In order to establish safe and effective use and disbursement of opioids, prescribers may require formal acceptance of an agreement between the provider and each patient to conform to a clearly defined set of expectations and restrictions guiding opioid therapy. Despite such pain agreements, some patients receiving opioid medications for the treatment of their cancer-related pain demonstrate a proclivity towards non-compliance with their agreement. This problem of noncompliance as abuse or diversion affects the broader community beyond the individual patient being treated and has gained attention from the media, the government, and society in general. Due to the potentially severe and far-reaching consequences of pain agreement violations in such instances, there is an obvious need for both the detection and prevention of the misuse of opioid medications. While this issue will persist as a challenge among providers caring for patients with cancer pain, any improvement in the recognition of individuals at higher risk for noncompliance with pain agreements can offer a clear benefit for early intervention. In an effort to establish any patient-related data that may assist with early recognition of those at higher risk for abuse, the association of patient demographics and noncompliance with pain agreements was examined among those treated at Roswell Park Cancer Institute’s (RPCI) pain clinic. This study investigated the incidence of and statistical relationships between pain agreement violations and demographic characteristics among a cross-sectional cohort of cancer pain patients at RPCI.
Materials and Methods
IRB approval and patient-signed consent were obtained. Urine toxicology screens and self-reporting among 283 individuals over a 24 month period were followed by monthly tests for 6 months. Demographic data including age, gender, race, marital and smoking status, history of EtOH or substance abuse, and psychiatric illness were compiled. Abuse encompasses positive testing for a non-prescribed opioid and/or any illicit substance. Diversion is absence of the prescribed opioid(s) in urine. Abuse rates were tabulated at all visits. Statistical analysis was performed in SAS v9.3 (Cary, NC) with 95% confidence intervals determined via the Agresti-Coull method and p-values less than 0.05 considered statistically significant. Associations between abuse, demographics, and visits were evaluated with Fisher’s exact and Chi-square tests.
Conclusions and Discussion
Over all visits examined in this study time period statistically significant associations were seen between smoking and abuse as well as between race and diversion, and the demographic characteristics of gender, marital status, smoking status, and a history of substance abuse were all independently associated with illicit substance consumption. At the first visits of the study time frame statistically significant associations were seen between smoking and abuse as well as between race and diversion, and the demographic characteristics of age, gender, marital status, smoking, and a history of substance abuse were all independently associated with illicit substance consumption. Most demographic characteristics examined in this study had some statistically significant association established between it and some form of noncompliance with the pain agreement. Only a history of alcohol abuse and a history of psychiatric illness were found to have no statistically significant association with pain agreement noncompliance. However, the specific form of noncompliance varied between demographic characteristics. Considering the statistically significant findings of this study, there is clearly the potential to identify specific demographic characteristics that may be associated with a higher risk of noncompliance with pain agreements in the forms of abuse or diversion. The elucidation of associations between such higher-risk characteristics and pain agreement violations may be beneficial for practitioners providing care for cancer pain patients by facilitating prevention, early detection, and/or intervention in an effort to reduce the impact of the opioid abuse and/or diversion epidemic. While clinical judgement and protocols will dictate the majority of a decision-making process regarding screening of individual patients, the addition of objective statistically significant data may assist with individualized screenings. The incidences of abuse and diversion were significantly higher in this study than observed in other non-cancer pain studies3. A possible explanation of this discrepancy is the referral to the RPCI pain clinic of patients who are either difficult to treat or exhibit aberrant behaviors. Therefore, these findings may not be representative of all oncology patients treated for cancer-related pain. While reproducibility with other studies is an important aspect of forming any clinical decision from investigational data, these findings suggest that patients treated at a specialized cancer pain clinic should be actively monitored for opioid abuse and diversion.