Painstaking Efforts to Improve Opioid Stewardship


Written By

Reviewed By


Liebschutz JM, Xuan Z, Shanahan CW, et al. Improving Adherence to Long-term Opioid Therapy Guidelines to Reduce Opioid Misuse in Primary Care: A Cluster-Randomized Clinical Trial. JAMA Intern Med. 2017;177(9):1265-1272.

Drug overdose is now the leading cause of death among Americans under the age of 50.1 This crisis has been building since the late 1990s, in part the result of rampant prescribing of opioid analgesics.2 In 2016, the United States (U.S.) Centers for Disease Control and Prevention (CDC) released a seminal guideline for primary care clinicians regarding opioid prescribing.3 These guidelines are now being implemented by clinicians, insurers, and healthcare institutions. The twelve recommendations contained in the CDC guidelines steer clinicians away from routine, long-term use of opioids.


While the guideline was developed using a systematic review methodology, meta-analyses were not conducted due to the low quality and high heterogeneity of included studies. Additionally, most of the recommendations were broad statements regarding patient management rather than the specific guidance clinicians expect. Despite these limitations, the guideline laid the foundation for shifts in institutional policy and governmental regulation that impact patients and providers.4 While none of the CDC recommendations directly addressed the use of patient-clinician agreements regarding opioid use (commonly referred to as pain contracts or controlled substance agreements), fulfillment of several recommendations might be achieved through that approach (Table 1).

Table 1. Select CDC Recommendations Pertinent to Patient-Clinician Agreements



2: Clinicians should establish treatment goals before prescribing opioids. Opioid therapy should only be continued if clinical improvements outweigh associated risks.

Very Weak

3: Clinicians should discuss risks and benefits of opioid therapy, as well as patient and clinician responsibilities, before prescribing opioids and periodically thereafter.


7: Clinicians should evaluate benefits and harms of opioid therapy every three months and within four weeks of initiation or dose escalation. Opioids should be tapered or discontinued if benefits do not outweigh harms.

Very Weak

8: Clinicians should periodically evaluate risk factors for opioid-related harms and incorporate strategies to mitigate them (e.g. dose tapering, naloxone co-prescribing).

Very Weak

10: Urine drug testing should be completed at baseline and at least annually.

Very Weak


The Transforming Opioid Prescribing in Primary Care (TOPCARE) study sought to assess the impact of a multicomponent care management intervention on opioid stewardship in four primary care centers.5 To achieve this goal, a randomized controlled trial was conducted with cluster randomization at the primary care clinician (PCC) level. The trial took place from January 2014 through March 2016 with outcomes evaluated at 12 months for each patient chronically prescribed opioids. The study sites included a large clinic affiliated with an academic health science center and three community-based federally-qualified health centers. Eligible PCCs were physicians or advanced practice nurses trained in family or internal medicine. They were also required to have at least four adult patients currently treated with long-term opioid therapy, defined as three opioid prescriptions at least 21 days apart in a six-month timeframe.


The intervention group received a multicomponent intervention consisting of nurse care managers (NCMs), an electronic patient registry, and academic detailing for PCCs. The NCMs performed initial and ongoing patient assessments for pain, addiction, and opioid misuse. A web-based electronic registry was used to facilitate population management by importing data from the electronic health record (EHR). Each PCC received a single 1-on-1 academic detailing session from an opioid prescribing expert. Both the intervention and control groups received access to electronic decision support tools via Tools available to both groups included pain assessments, substance abuse assessments, and links to state prescription monitoring programs, among other resources. The primary outcome of the study was the proportion of patients achieving “guideline-concordant care”, defined as execution of a patient-clinician agreement for opioid use and a urine drug test (UDT) during the 12-month observation period. Secondary outcomes included opioid discontinuation, opioid dose reduction of at least 10%, and the frequency of early refills.


Baseline characteristics were similar between groups with a few notable exceptions. Patients in the control group were significantly more likely to have a drug use diagnosis, a mental health diagnosis, and be non-English speaking. There was no notable difference at baseline in the mean morphine-equivalent daily dose (MEDD) prescribed by PCCs. The typical patient in this trial was a 50-60 year old white, English-speaking, male or female with a mental health diagnosis taking a mean MEDD of 61.5mg.


Analysis of the primary and secondary outcomes at 12 months revealed significant differences in nearly every parameter favoring the intervention (Table 2). Patients whose PCCs had received the multicomponent intervention were more likely than patients assigned to control PCCs to achieve “guideline-concordant care” with both a patient-clinician agreement and a UDT in the past 12 months. Similar differences were observed for each component of the primary outcome analyzed individually. Differences were also observed in secondary outcomes including opioid discontinuation rates and opioid dose reductions >10%. While there was no significant difference between groups at 12 months, the mean MEDD increased slightly in the control group while remaining stable in the intervention group.


Table 2. Primary and Secondary Outcomes


Intervention (n=586)

Control (n=399)

P Value

(95% CI)

Patient-clinician agreement AND

urine drug test within past 12 months

386 (65.9%)

151 (37.8%)


3.3 (1.9-5.6)

Patient-clinician agreement

489 (83.5%)

243 (60.9%)


2.5 (1.4-4.5)

Urine drug test within past 12 months

437 (74.6%)

231 (57.9%)


2.4 (1.3-4.4)

Opioid discontinuation

125 (21.3%)

67 (16.8%)



Opioid dose reduction ≥ 10%

151 (32.8%)

76 (22.9%)



Opioid discontinuation OR

opioid dose reduction ≥ 10%

276 (47.1%)

143 (35.8%)



MEDD in mg - mean (SD)

60.8 (93.7%)

67.3 (80.4%)




This trial demonstrated that a multicomponent intervention with academic detailing for PCCs, a devoted opioid case manager, and an electronic patient registry can increase adherence to commonly recommended opioid stewardship strategies. A thorough assessment of study methodology supports the conclusion that the primary driver of this improvement was likely the care management component. The trial included two centrally located (off-site) NCMs shared between 25 PCCs with an average of 23 patients each.6 This equates to approximately 290 patients per NCM resulting in roughly ten patients per day to manage, assuming monthly contact. It is also possible that the academic detailing component of the intervention positively affected outcomes. A previous study demonstrated that training 27 PCCs to order UDTs annually and select appropriate diagnosis codes for pain management influenced behavior when combined with EHR support and financial incentives.7 Meanwhile, the relatively poor results in the control group highlight the inadequacy of offering just decision-support tools.


While the intervention was successful in promoting some tenets of opioid stewardship, it does not provide direct proof that the intervention results in decreased patient harm.  The study lacked patient-oriented outcomes, such as pain control, quality of life, and opioid-related overdoses.  Therefore, it is not clear whether dose reductions and treatment discontinuations were appropriate. The results of the NCM assessments were not reported. While the NCM’s utilized prescription monitoring programs (PMPs), the findings from the PMPs were not reported. Further, the use of non-opioid medications for chronic pain was not quantified or evaluated.  Lastly, the methodology for detecting early refills was too flawed to make any conclusions for that outcome.


It is not a surprise that devoting significant resources (e.g. academic detailing and case management) led to higher rates of patient-clinician agreements and UDT.  In the absence of patient-oriented outcomes it is unclear if this study justifies allocating these resources. The improved likelihood of opioid dose reduction and discontinuation is compelling, but while higher doses of opioids are clearly associated with increased risk of opioid-related harm there is little evidence demonstrating that broadly-applied opioid dose reductions decrease patient harm. Future research of this care model must include assessment of patient-oriented outcomes to inform responsible opioid stewardship.


The TOPCARE website offers little information specific to pharmacy practitioners. However, ambulatory care pharmacists can (and should) play a role in this model of care. For example, a clinical pharmacist specializing in pain management could conduct the same assessments as the NCMs while also actively manage opioids and other analgesics under a collaborative practice agreement. Further, the addition of a community pharmacist component could facilitate information gathering regarding early refills and risk for opioid harm based on a thorough assessment of concurrent medication use. Pharmacists can also provide naloxone. Coupled with patient and caregiver education, naloxone is a life-saving intervention that was not discussed in this study (see: Opioid Safety & Overdose Prevention Resource and Naloxone Coprescribing Commentary).


Based on the TOPCARE study results, it seems reasonable for primary care practitioners and practice groups to implement systematic opioid stewardship programs. However, patients’ pain must also be adequately addressed alongside careful consideration of opioid-related risks. An ambulatory care pharmacist providing comprehensive medication management could implement this service. But do ambulatory care residencies provide sufficient experience managing chronic pain to prepare graduates for this role? To rise to this challenge, all ambulatory care and community pharmacy residencies should offer a pain management experience and opioid harm reduction education.


How are you contributing to opioid stewardship efforts? Does residency training provide sufficient experience? What barriers have you encountered? Let us know – write a comment and check out our podcast!