Setting Expectations, Following Orders, Safety, and Standardization: Clinicians’ Strategies to Guide Difficult Conversations About Opioid Prescribing

Abstract

Background

Evidence has continued to accumulate regarding the potential risks of treating chronic pain with long-term opioid therapy (LTOT). Clinical practice guidelines now encourage clinicians to implement practices designed to reduce opioid-related risks. Yet how clinicians implement these guidelines within the context of the patient encounter has received little attention.

Objective

This secondary analysis aimed to identify and describe clinicians’ strategies for managing prescription opioid misuse and aberrant behaviors among patients prescribed LTOT for chronic pain.

Design

Individual interviews guided by a semi-structured interview protocol probed: (1) methods clinicians utilize to reduce prescription opioid misuse and address aberrant opioid-related behaviors; (2) how clinicians respond to misuse; and (3) resources and constraints faced in managing and treating misuse among their patients.

Participants

Interviews were conducted with 24 physicians and nurse practitioners, representing 22 Veterans Health Administration (VA) facilities across the USA, who had one or more patients in their clinical panels who were prescribed LTOT for the treatment of chronic non-cancer pain.

Approach

Qualitative content analysis was the analytic approach utilized. A codebook was developed iteratively following group coding and discussion. All transcripts were coded with the finalized codebook. Quotes pertaining to key themes were retrieved and, following careful review, sorted into themes, which were then further categorized into sub-themes. Quotes that exemplified key sub-themes were selected for inclusion.

Key Results

We detail the challenges clinicians describe in navigating conversations with patients around prescription opioid misuse, which include patient objection as well as clinician ambivalence. We identify verbal heuristics as one strategy clinicians utilize to structure these difficult conversations, and describe four heuristics: setting expectations, following orders, safety, and standardization.

Conclusion

Clinicians frequently use verbal heuristics to routinize and increase the efficiency of care management discussions related to opioid prescribing, redirect responsibility, and defuse the potential emotional charge of the encounter.

INTRODUCTION

As evidence has accumulated regarding potential harms of treating chronic pain with long-term opioid therapy (LTOT),^{1, 2} clinical practice guidelines have encouraged practices designed to reduce the risks of opioid-related harms.^{3, 4} Recommended practices include activities such as use of state prescription drug monitoring programs (PDMPs); educating patients about benefits and harms of opioid therapy, and outlining conditions for on-going opioid prescribing, including routine urine drug testing (UDT); and behavioral monitoring—with dose reduction, tapering, or discontinuation of opioid therapy when the risks of continued prescribing outweigh the benefits. These efforts have contributed to a reduction in prescription opioids for the treatment of chronic pain and a decline in prescribing practices associated with adverse outcomes.^{5, 6} While the reason for these changes to clinical guidelines and practice is evident, how clinicians incorporate new treatment approaches into clinical care, and do so without disrupting the clinician-patient relationship, has received little attention.

Clinicians are now tasked with engaging in conversations about care management practices around prescription opioids. These conversations add to the challenges faced by clinicians caring for patients with chronic pain.^7–9 Further, both patients and clinicians experience tensions and frustration regarding pain care,¹⁰ with clinicians dissatisfied with the pain care they deliver,^{7, 11} and patients dissatisfied with care they receive.¹²

As clinical guidelines have changed, research has begun to address the dynamics of clinician-patient communication regarding opioid prescribing. Prior research has identified themes identified by patients as preferred for communication processes around opioid tapering¹³ and their experiences undergoing opioid taper.¹⁴ Less is known about clinicians’ perspectives of conversations with patients regarding opioid management. Drawing upon interviews with clinicians caring for patients prescribed LTOT for chronic pain, we detail the challenges they describe in navigating conversations with patients around opioid misuse, and identify strategies they utilize to structure and manage these conversations.

METHODS

This study is a secondary data analysis of a larger study focused on barriers to UDT among patients prescribed LTOT for chronic pain. The goal of this analysis was to understand how clinicians adhere to recommendations for managing patients prescribed LTOT. The interview guide was developed by clinician-researchers with expertise in the treatment of chronic pain, LTOT, substance use disorders, and qualitative research methods. Feedback on the interview guide was solicited from a multidisciplinary team specializing in the treatment of chronic pain in primary care. The Institutional Review Board at the VA Portland Health Care System approved the study and procedures. All participants provided written informed consent to participate.

Recruitment

Eligible participants were VA clinicians with one or more patients prescribed LTOT for chronic pain. We queried administrative databases to identify eligible clinicians. Participants were initially contacted with a recruitment letter summarizing the purpose of the study. Those who indicated interest were sent informed consent materials and contacted by phone to discuss the scope of the study. All study visits were completed by telephone. We recruited interview participants and conducted interviews until we had achieved data saturation for the primary study focus (UDT), the point at which new interviews began to yield few additional insights. Interviewing took place between March 2015 and May 2016. Our final sample was composed of 24 VA clinicians representing 22 VA Medical Centers across the USA.

Data Collection

All interviews were conducted by the project investigators (L.G. or B.J.M.), lasted 30–40 min, and audio-recorded. Interviews were guided by a semi-structured interview, which probed (1) methods clinicians utilized to reduce prescription opioid misuse and address aberrant opioid-related behaviors; (2) how clinicians responded to misuse; and (3) resources and constraints they faced in managing and treating opioid misuse among their patients. Participants were offered a $50 store gift card for participating, provided the interviews were conducted outside of their usual work hours. All data were deidentified and transcribed verbatim.

Data Analysis

We used a qualitative content analysis approach.¹⁵ Six interviews were coded jointly by project investigators to establish mutually agreed upon codes and definitions, which were then used to build a codebook. The remaining interviews were divided and first coded independently by project investigators, and then exchanged for secondary coding (i.e., all interviews were coded by two investigators). Study investigators met to discuss areas of divergent coding and came to agreement on all codes in all transcripts. AtlasTI version 7 was used to organize and code data. Quotes pertaining to conversations between patients and clinicians were retrieved and sorted into themes, which were then further categorized into sub-themes. Quotes that exemplified key sub-themes were selected for inclusion in this manuscript.

RESULTS

Participants

The 24 respondents were physicians (83%) or nurse practitioners (17%), with most specializing in internal medicine (46%) or family medicine (12.5%). Other clinicians were specialists in geriatric medicine, psychiatry, addiction medicine, and physical medicine/rehabilitation. On average, interview participants were 49.5 (SD = 10) years of age, and 63% were female. The average number of years since completion of training was 17 (SD = 10), with a range of 2–37. Participants identified as white (67%), Asian (21%), and biracial (8%). Our interview sample included clinicians representing diverse geographic regions: 2 were located in the northeast, 4 in the South/mid-Atlantic, 6 in the northwest, 8 in the southwest, and 4 in the mid-west.

Themes and Sub-themes

In analyzing the data regarding clinician-patient conversations around LTOT prescribing, a primary theme was clinicians’ struggles to navigate and successfully manage these conversations. Respondents described two challenges in particular: patients objected to changes in prescribing, and clinicians experienced ambivalence in altering their practice to conform to the new guidelines. The second primary theme that emerged was clinicians’ use of verbal heuristics, or short-cuts, to structure and guide these difficult conversations. Clinicians described using these strategies to facilitate problem solving and increase efficiency, particularly within the context of ambiguous or complicated situations. Primary themes, sub-themes, and representative quotes are included in Table 1.

Table 1.

Summary of themes and sub-themes

Theme	Sub-theme	Example
(1) Difficult conversations
	(a) Patient objections and complaints	“they beg, they plead, they think if they talk to you enough you’ll change your mind… they go to the patient advocate and complain.”
	(b) Clinician ambivalence	“…it’s very hard to apply the new feelings on this to people who have been managing a different way for a very long time and I worry that it’s a little unfair to patients to all of a sudden…”
(2) Clinician strategies: verbal heuristics for difficult interactions
	(a) Safety heuristic	“Okay, it’s clear to us that you are not following through with the guidelines of the contract. And if that’s the case then… I do not feel comfortable prescribing for you anymore because you are using in a way that’s unsafe.”
	(b) Setting expectations heuristic	“I establish ground rules with them and now I am even saying no early refills even for legitimate reasons…”
	(c) Following orders heuristic	“I try to act as if this is just some kind of big cog in the government wheel and there’s nothing I can do.”
	(d) Standardization heuristic	“I make it a point to say that I do this for everybody so I that do not forget to do it on anybody…I do it for all my patients who are on prescription opioids whether they are 29 or 85…”

Difficult Conversations

Clinicians identified both aspects of patient behavior and their own thoughts and beliefs that contributed to difficulty in conversations around opioid management.

1a. Patient Objections and Complaints

Clinicians reported that monitoring, testing and, when necessary, tapering or discontinuing opioid prescriptions significantly stressed the clinician-patient relationship. Indeed, one participant described initiating an opioid taper for a new patient as, “often very relationship destroying.” Confrontational, emotionally charged conversations could ensue as Veterans were faced with a new approach to prescribing:

…a lot of these patients have been on it for a very long time and they are very resistant to tapering themselves off and…that is definitely a big, big barrier…we have to spend a lot of time with the patients explaining the risks and still they resist and they feel that they are being denied their right or care, medical care. “I was told I was going to be taken care of and you’re not taking care of me” “this is one that helps me, this is my life” “you made me go through all this and I’m in pain…”

Clinical guidelines and local standards for opioid prescribing have changed rapidly, and clinicians found themselves in the uncomfortable position of implementing this practice change with patients who had become accustomed to receiving opioids for chronic pain.

Conversations about opioids were also frequently mentioned as one of the most time-consuming aspects of clinicians’ jobs. As one clinician noted “Oh the time to address and withdraw narcotics…It just takes hours.” Another noted relief that resulted from group education visits: “nursing and clinicians do one big group visit to talk with patients about the risks… so that it isn’t a process we have to go through person by person, which is very time consuming and…eats up our access.” In addition, these interactions were often unpleasant. As one clinician stated, “they’re [patients] not happy about it and they let us know that.” In other cases, patient reaction could be more extreme; patients could be angry, aggressive, and even violent in reaction to clinicians’ changes to their opioid prescriptions:

…I’ve had someone hit me with their cane. I’ve had my car keyed... I already had someone that wanted to kill me several years ago about this…

Objections were not just voiced with clinicians, complaints were also frequently shared with patient advocates or hospital administration. As one clinician described, “they beg, they plead, they think if they talk to you enough you’ll change your mind… they go to the patient advocate and complain. And then the patient advocate, you’ve got to defend yourself to…” Other clinicians described the implications of patient complaints to congressional officials, a practice mentioned across multiple interviews.

Finally, clinicians found it difficult to be on the receiving end of complaints regarding their perceived lack of concern for patients’ pain, when they believed that their actions were ultimately in the patient’s best interest. As one clinician stated, “… a majority of the people I work with got into this profession because we wanted to help people and…make a positive impact and, you know, we’re kind of softies at heart.” Indeed, multiple respondents described feeling “bad” or “alone” in their attempts to implement guideline-recommended opioid prescribing practices with patients on LTOT.

1b. Clinician Ambivalence

Challenges clinicians faced in their communication with patients prescribed LTOT stemmed not just from negative patient reactions to these conversations, but also their own ambivalence about enacting guideline-recommended changes. Although clinicians recognized that LTOT was associated with heightened risk for patients on a population-level, applying this knowledge to individual patients could feel uncomfortable, as one clinician described:

It’s a different beast with your long-term patients who have been on high dose opioids and have been using them regularly for the last twenty years without event…I mean the standards have changed, and changed rapidly, which I think is great and I really support, but it’s very hard to apply the new feelings on this to people who’ve been managing a different way for a very long time and I worry that it’s a little unfair to patients to all of a sudden…

While this clinician may have believed that the new clinical approach was positive, enacting changes to patients’ prescriptions nonetheless felt difficult. Another clinician described facing a similar challenge in disallowing the use of concurrent marijuana and opioids, even though the patient had always been forthright about using both substances:

With a lot of patients I inherited who have been on marijuana for years, who have been very open about their marijuana use, who have been on chronic opiates for years who have already seen pain clinic and pain clinic says we don’t recommend concurrent use of opiates and marijuana. But they’ve been on it forever—I have a very hard time giving them a good reason why [I] should then cut them off now.

In sum, clinicians found conversations about guideline-recommended opioid prescribing practice changes to be challenging. While clinical guidelines and new information about prescribing practices encouraged clinicians to taper high doses of opioids, particularly among patients evidencing aberrant behaviors, some patients resisted these changes in ways that were emotionally taxing and time-intensive for clinicians. Adding to the complexity, clinicians could themselves be ambivalent about enacting guideline-recommended changes.

Clinician Strategies: Verbal Heuristics for Difficult Interactions

To manage difficult conversations, clinicians shared what we term “verbal heuristics”—essentially a pre-packaged response or conversational short-cut—to more quickly and efficiently guide and defuse challenging, emotional conversations. Heuristics were frequently utilized to describe the purpose of urine drug screening, and also employed when clinicians planned to discontinue or taper opioids following a patient’s aberrant behaviors. We identified four varieties of heuristics among clinicians, which we categorized as verbal heuristics of safety, setting expectations, following orders, and standardization. Although we distinguish these heuristics for analytic purposes, strategies may evidence some degree of conceptual overlap in practice.

2a. The Safety Heuristic

Safety was a common verbal heuristic participants relied on to guide conversations. One participant described a typical phone conversation she might have with a patient following a second aberrant UDT as the following, “Okay, it’s clear to us that you’re not following through with the guidelines of the contract. And if that’s the case then… I don’t feel comfortable prescribing for you anymore because you’re using in a way that’s unsafe.” In this heuristic, the clinician first brings attention to the signed informed consent (which she refers to as a “contract”) and the patient’s violation of it (redirecting attention from the clinician’s decision to the patient’s choices), and then states that concerns about patient safety preclude her from continuing to prescribe opioids for the patient.

Another clinician tied the safety heuristic to her own responsibilities as a physician, “when I became a doctor I took an oath to do no harm and if…you’re using illicit substances in addition to your pain medications, or… getting twice as many pain medications…as I thought, then we’re running a very serious risk of doing harm and…that’s unethical for me.” Like other clinicians, she used the safety heuristic to redirect responsibility for an unwelcome decision from herself to the patient. She explicitly voiced patients’ responsibility for the decision by employing a phrase she had learned from a colleague, “Your choices are limiting my options.”

2b. The Setting Expectations Heuristic

Participants also underscored the importance of setting expectations regarding adherence to the treatment plan. As one clinician stated, “I establish ground rules with them and now I am even saying no early refills even for legitimate reasons…I joke with them maybe if there [is] a direct nuclear explosion on your house I might allow that as a legitimate excuse.” This clinician believed that instilling the expectation with patients that prescribing practices would not be flexible helped curtail difficult conversations in the future. Others described the importance of responding firmly and consistently to aberrant behaviors, as one clinician described, “If they violate my contract, I stop prescribing to them.” While some clinicians describe tolerating a few infractions, “I let people have a learning curve and give them some room,” others describe a firmness and lack of flexibility, “I offer them alternatives, pain management medications, and strategies, but I do not debate the [UDT] results.”

2c. The Following Orders Heuristic

Participants also described a verbal heuristic we classified as following orders to ease tense conversations. In this strategy, clinicians claim that opioid prescribing decisions were not their own, but determined by VA policy, “I try to act as if this is just some kind of big cog in the government wheel and there’s nothing I can do.” This clinician felt that using language suggesting that clinical decisions were determined institutionally would make patients less likely to challenge her decision to taper or discontinue opioid therapy following an aberrant UDT. This strategy often leveraged the opioid informed consent document clinicians discuss with patients before initiating a long-term opioid prescription.¹⁶ Clarifying possible repercussions through a signed informed consent made consequences of aberrant behaviors “…pretty clear from the get-go,” as one participant described. When patients evidenced aberrant behaviors detailed in the informed consent, some clinicians indicated that they would review this language with the patient when discussing a decision to taper or discontinue:

…if it’s an illegal substance like cocaine, I just review the drug screen results with the patient. I say that…the patient has already signed the pain agreement in the past. I mean, we make sure that every, at least every two years we sign a pain agreement. And we go over it, the wording of the pain agreement that they agreed not to use any illegal substances…And we say that we have no choice than to… stop the pain medications.

This approach highlights the patient’s role in initiating the discontinuation, presenting it as a natural consequence of not adhering to the rules. Further, the clinician suggests that the decision is not her own, but rather determined by language in the informed consent document. Yet notably, the language of the informed consent leaves the decision to taper or discontinue up to the clinician, stating “If you do not take opioids responsibly, your clinician may stop your prescription.” Thus, while framed as a necessary response based on the consent, the decision ultimately remains the clinician’s. Similarly, another clinician stated, “And we go over it, the wording of the pain agreement that they agreed not to use any illegal substances while using…opioids. And we say that we have no choice than to …stop the pain medications.”

Deflecting responsibility, other clinicians described their unwelcome prescribing decisions as necessitated by decisions made by their hospital’s opioid safety committee. As one clinician described his process, he would request an “opioid flag” for a patient, which would then initiate a patient review with the opioid safety committee. Then, “we can send them a letter to tell them why we are taking them off, the whole (opioid) committee has met and we have made this decision to take you off because you have illegal substances and you have unsafe behaviors.” Likewise, another clinician explains “this committee of specialists, they have reviewed your case and this is what they are telling me what is recommended and I need to follow these recommendations.”

2d. The Standardization Heuristic

Finally, a standardization heuristic was described, in which clinicians worked to normalize the use of UDT by explaining that it was a standard of care for patients prescribed LTOT:

I make it a point to say that I do this for everybody so I that don’t forget to do it on anybody…I do it for all my patients who are on prescription opioids whether they’re 29 or 85… it’s sort of like… checking an A1C in a diabetic… Something that is part of a routine monitoring for… this condition and this medication.”

Similarly, another clinician described emphasizing UDT as a routine aspect of clinical care in conversation with patients, “I like to say it, specifically say it so that’s it’s not a surprise and specifically tell them why I’m doing it because I’m going to do this for everyone at least once a year.” For these clinicians, the standardization heuristic worked to reassure patients, that UDT was in fact routine, universal, and non-discriminatory.

DISCUSSION

Interviews with clinicians treating patients prescribed LTOT for chronic pain identified challenges they faced in navigating conversations about opioid management, stemming from patient dissatisfaction and clinician ambivalence about enacting guideline-recommended changes. Many of the challenges we identified mirror those described by clinicians caring for patients with chronic pain and opioid prescriptions generally,^{8, 11} as well as initiating opioid tapers specifically.¹³ We identified verbal heuristics as one strategy clinicians utilize to ease conversations about opioid prescribing. Verbal heuristics were used by clinicians to routinize and increase the efficiency of the discussion, redirect responsibility away from the clinician, and defuse the potential emotional charge of the encounter.

One question that arises from these findings is the extent to which these strategies align with a patient-centered approach to care, as recommended by the Institute of Medicine¹⁷. Patient-centered care encourages clinicians to prioritize the therapeutic alliance, listen and understand the patient’s experience, express empathy, and involve patients in shared decision-making.¹⁸ Clinicians caring for patients with chronic pain often experience greater satisfaction when trained in a shared-decision-making care model,^{19, 20} and shared decision-making may improve clinician-patient interaction.²¹ As described here, relying on a heuristic of following orders, for example, to justify opioid tapering or discontinuation decisions may not allow for the patient input, personal tailoring, and shared decision-making recommended in a patient-centered approach. Rather, this strategy seems designed to deflect and defuse, and may not allow patients to feel they are a partner. Conversely, attributing decision-making to an entity outside of the clinician-patient dyad may facilitate both clinician’s communication and patient’s experience of empathy, which could enhance the therapeutic relationship in the context of this otherwise potentially contentious issue.²²

Heuristics of safety and standardization, in contrast, may provide a better conversational starting point for patient-centered care. The safety heuristic highlights the concern for patient well-being, while the standardization heuristic equates opioid management practices with standard care practices used to monitor other medical conditions. Both approaches hold potential to strengthen the therapeutic alliance—by expressing care and normalizing practices associated with opioid prescribing. These approaches also align with strategies identified in prior research. For example, communication may be enhanced when patients are offered individualized reasons for tapering, clinicians express empathy through the process, and opioid tapers are designed with patient input.¹³ Safety messages have also been shown to be better received by patients prescribed LTOT than rule-following messages.²³ Finally, it has been recommended that clinicians strive to replace a law enforcement or deal-making approach to opioid prescribing focused on patient behavior—where clinicians aim to “catch” patients or negotiate with them—with a framework that weighs benefits to harms of the medication.²⁴

Future research could evaluate how patients experience the verbal heuristics detailed here, and identify those perceived to be least relationship-challenging, as well as those associated with patient engagement and positive health outcomes over time. For instance, if a verbal heuristic of following orders is linked with a higher rate of patient disengagement in care, clinicians could be encouraged to utilize more effective alternatives. Such efforts could then ground the verbal heuristics clinicians utilize in evidence, rather than anecdote. Additional strategies to ease tense conversations about opioid safety that fall within a patient-centered, shared decision-making framework should also be identified and similarly tested. Future research is also needed to replicate study findings.

In these interviews, clinicians rarely indicated that they discussed addiction with patients. This could reflect several factors, including patient rejection of such a characterization of their opioid use (as patients generally endorsed chronic pain, rather than substance use disorder), clinician hesitancy to frame opioid misuse as addiction, and/or clinician discomfort with diagnosing addiction in their patients. When the subject of patient addiction was broached, it was generally in the context of referring patients tapered or discontinued from LTOT to a specialty substance use disorder treatment program.²⁴ Clinicians may benefit from additional guidance and support regarding how to approach the subject of addiction with their patients in this context. Finally, this research aligns with prior work suggesting that clinicians would benefit from support as they transition to new standards of opioid care,^{8, 9} as well as eased access to or additional resources for pain management approaches that do not involve opioids.²⁵ Clinician peer support groups have been suggested as one promising approach.^{8, 26} More generally, decline in LTOT use for pain should be paired with increased availability of non-opioid pain treatments with demonstrated evidence in improving pain, function, and quality of life.^27–30 Expanding access to such resources could help reduce clinicians’ ambivalence about prescription opioid dose reduction or discontinuation, and bolster guideline-recommended pain care practices.

Limitations

This research is qualitative and not intended to speak to the frequency with which verbal heuristics are utilized, nor how representative such strategies are in clinician-patient interactions. Second, this manuscript details clinicians’ beliefs and perspectives around conversations about opioid prescribing; patients may view these conversations differently. Third, all participating clinicians were providing services in VA hospitals and clinics, and as such the findings may not to other settings. Fourth, these results are a secondary analysis of interviews where the primary focus was on addressing barriers to UDT; study findings may differ if the focus were exclusively on identifying clinician strategies for managing difficult conversations. Finally, study results represent clinician perceptions and there are other external issues impacting prescription opioid safety and management.

CONCLUSION

This research identified challenges clinicians face in navigating conversations about opioid prescribing. While difficulties arose from patient objection, they also stemmed from clinicians’ ambivalence about implementing the changes to care recommended by clinical practice guidelines. We described the use of verbal heuristics—setting expectations, following orders, safety, and standardization—clinicians utilized to structure these difficult conversations. As evidence emerges regarding the potential risks of LTOT and limited benefits,^{3, 31} clinicians may continue to face pressure to increase monitoring of patients’ opioid use and to taper and discontinue opioids for patients evidencing high-risk behaviors. Yet these clinical interactions come with substantial costs: they can be time-intensive, conflictual, and potentially undermine the clinician-patient relationship. Given these dynamics, research identifying effective strategies to guide difficult conversations around opioid safety are needed.

Funding

Research reported in this manuscript was financially supported by an award from the US Food & Drug Administration (FD004508). The work was also financially supported by resources from the VA Health Services Research & Development-funded Center to Improve Veteran Involvement in Care at the VA Portland Health Care System (CIN 13-404).

Compliance with Ethical Standards

The Institutional Review Board at the VA Portland Health Care System approved the study and procedures. All participants provided written informed consent to participate.

Conflict of Interest

The authors declare that they do not have a conflict of interest.

Disclaimer

The views expressed in this article are those of the authors and do not reflect the position or policy of the Department of Veterans Affairs, the US Food & Drug Administration or the US government.