Sep-Oct_2015_CoverDear Editor:

An estimated 75 million Americans experience nonmalignant chronic pain,[1] and oftentimes the responsibility for management falls to the primary care provider. Therefore, pain management is an increasingly important clinical skill for residents in primary care fields to master. The use of a narcotic agreement or contract may be one such skill, with reported rates of use in resident clinics varying from 10 percent to 82 percent.[2,3] However, utilization of narcotic agreements to mitigate the misuse of controlled substances remains somewhat controversial in the literature (i.e., questionable efficacy).[4–6] In this study, we examined whether a change in clinic policy from recommended narcotic agreements (Time-1) to required narcotic agreements (Time-2) would improve resident contracting and/or have any effect on patient behavior (as measured by provider requests for urine drug testing) in an internal medicine resident-provider outpatient clinic (30 providers).

Participants consisted of 299 patients, 223 of whom provided data at Time-1 (during the period of recommended narcotic agreements) and 76 of whom entered the practice during Time-2 (after the initiation of required narcotic agreements). Of the 223 patients seen during the Time-1 period, 106 (47.5%) were also seen during the Time-2 period. So, the Time-2 sample consisted of 182 patients (106 who were also included at Time 1 plus 76 new patients). Of the 299 patients, 122 (40.8%) were male and 177 (59.2%) were female. The age of the sample ranged from 23 to 82 years (M [standard deviation, SD]=50.15 [11.36]). As for marital status, 123 (41.1%) participants were single, 79 (26.4%) divorced, 75 (25.1%) married, and 22 (7.4%) widowed. With regard to employment status, 110 (36.8%) were unemployed, 82 (27.4%) employed/in-school, 60 (20.1%) disabled, and 47 (15.7%) retired. As for prescribed narcotics, 77 (25.8%) participants were prescribed a hydrocodone combination, 58 (19.4%) tramadol, 41 (13.7%) an oxycodone combination, and the remainder “other.”

Using a retrospective approach (chart review), we identified all outpatients being prescribed chronic narcotic therapy (defined as two temporal prescriptions for narcotics within a six-month time period) either before or after an index appointment during Time-1 (May–September 2014) or Time-2 (January–March 2015). We then compared participants during the two time periods with regard to 1) signed narcotic agreements and 2) documented urine drug testing (typically requested because of suspicious patient behavior). We excluded the months of October through December of 2014 to allow implementation time for the new policy regarding the narcotic agreement (i.e., from “recommended” to “required”). If the patient had signed a narcotic agreement before implementation of the new policy, he or she was asked to sign the new agreement after the implementation date.

At Time-1, 113 (50.7%) of the 223 patients were categorized as chronic users of prescribed narcotics. Of these 113 patients, 27 (23.9%) had signed a narcotic agreement and 67 (59.3%) had a documented urinary drug test. At Time-2, 150 (82.4%) of the 182 patients were categorized as chronic users of prescribed narcotics. Of these, 150 patients, 89 (59.3%) had signed a narcotic agreement and 74 (49.3%) had a documented urinary drug test. Note that the percentage of chronic narcotic-use patients who signed a narcotic agreement was statistically significantly greater at Time-2 compared to Time-1, chi2=32.83, p<0.001. However, the percentage of chronic narcotic-use patients who had a documented urinary drug test was not statistically significantly different at Time-1 compared to Time-2, chi2=2.57, p<0.11. In addition, of the 27 chronic narcotic-use patients at Time-1 who signed a narcotic agreement, 18 (66.7%) remained at Time-2, whereas of the remaining 86 chronic narcotic-use patients at Time-1 who did not sign a narcotic agreement, 56 (65.1%) remained at Time-2 (no statistically significant difference, chi2=0.33, p<0.59).

Findings indicate that two-thirds of the patients who were prescribed chronic narcotic therapy during Time-1 were still present during Time-2. The absent one-third during Time-2 may have been due to factors such as geographic relocation, resolution of pain, “doctor shopping,” finances, or changes in insurance. In comparing the two study periods, only 23.9 percent of patients signed the narcotic agreement at Time-1, whereas 59.3 percent signed the narcotic agreement at Time-2—a statistically significant difference, which indicates that the policy change improved contracting tenacity by the residents. However, in comparing the two time periods with regard to urine drug tests, which are typically ordered for suspicious patient behaviors, there was not a statistically significant difference (59.3% vs. 49.3%). Findings suggest that while residents improved in their ability to obtain signed narcotic agreements when providing prescriptions for chronic narcotic therapy, there was seemingly no change in suspicious behavior by the patients.

There are a number of potential limitations in this study, including the small sample size; use of a resident clinic with a high percentage of indigent patients (potential limitations with regard to generalizing findings to other types of populations); comparison of unequal time periods in terms of months (but comparable subject numbers); documentation of urine drug testing, which reflects completed not requested urine drug testing; and some urine drug tests during Time-2 may have been ordered without suspicion as the new contract states that up to two are required per 12 months (however, Time-2 was only three months in duration, and perfunctory urine drug tests were less likely). Regardless, this is the first study to our knowledge to examine the effects of a policy change on narcotic agreements with patients on chronic narcotic therapy in a resident-provider clinic. Findings indicate that while the percentage of signed narcotic agreements obtained by residents was meaningfully improved by requiring contracting versus recommending contracting, the prevalence of drug testing (oftentimes ordered because of suspicious patient behavior) did not appear to change.

References.

1. Hariharan J, Lamb GC, Neuner JM. Long-term opioid contract use for chronic pain management in primary care practice. A five year experience. J Gen Intern Med. 2007;22(4):485–490.
2. Subedee A, Shrestha DP, Olugbodi A, et al. Improving chronic opiate management in a resident clinic: a QI project. Presented at the Pennsylvania Eastern 2010 Abstract Competition; 2010 November.
3. Yanni LM, Weaver MF, Johnson BA, et al. Management of chronic nonmalignant pain: a needs assessment in an internal medicine resident continuity clinic. J Opioid Manag. 2008;4(4):201–211.
4. Miller G. Opioid agreements: helping or harming the physician-patient relationship? Pain Med News. 2011;9:3.
5. King SA. How useful are patient opioid agreements and urine drug testing? Psychiatr Times. March 2, 2011.
6. Collen M. Opioid contracts and random drug testing for people with chronic pain–think twice. J Law Med Ethics. 2009;37(4):841–845.

With regards,
Melissa M. Butler, MD; Randy A. Sansone, MD; and Charles Opperman, MD
Dr. Butler is Clinical Assistant Professor, Department of Internal Medicine, Wright State University School of Medicine, Dayton, Ohio, and Assistant Medical Director, Sycamore Primary Care Center, Miamisburg, Ohio; Dr. Sansone was Professor, Departments of Psychiatry and Internal Medicine at Wright State University School of Medicine in Dayton, Ohio, and Director of Psychiatry Education, Kettering Medical Center, Dayton, Ohio at the time of this study; Dr. Opperman is a resident in the Department of Internal Medicine, Kettering Medical Center, Dayton, Ohio.

Funding/financial disclosures. No funding was received for the preparation of this article. The authors have no conflicts of interest relevant to the content of this letter.