Addressing prescription opioid abuse concerns in context: synchronizing policy solutions to multiple complex public health problems.
INTRODUCTION I. THE MISUSE OF OPIOIDS: REALITIES AND MINIMIZED CONCERNS A. Rises in Prescriptions and Opioid Related Injuries: An Illusory Correlation? B. Broad Definitions and Tunnel Vision in Nonmedical Use II. PAIN--ONE OF MULTIPLE INTERACTIVE, DYNAMIC, COMPLEX PUBLIC HEALTH PROBLEMS A. Suicidality B. Mental Illness C. Substance Use Disorders D. Providers' Obligations to the Patient & Self-regulation through Virtue III. DECISION-MAKING ERRORS BY PROVIDERS & POLICYMAKERS A. Social Cognitive & Moral Decision-making B. Dual Process Models of Decision-making C. Biases & Heuristic Failure in Decision-Making Errors IV. REGULATION, POLICY, & PRACTICE: RESPONSES FROM INCOHERENT TO RATIONAL A. Regulating Prescribing at Federal Level B. Regulating Prescribing and Practice at the State Level V. POLICIES THAT REDUCE HARM WHILE PRESERVING GOODS A. Safe Storage & Disposal B. Naloxone Access, Distribution, and Immunity C. Conclusions
"[A]n ethically and clinically sound public policy response will not discount or disfavor one crisis in order to obsessively focus upon the other." (1)
Hyrum Neizer would be dead today if his wife hadn't walked in on him with the gun in his mouth. He was in unremitting pain ... his life was ruined not just by his chronic physical pain but by the very people who were supposed to be helping him. ... They eventually made him believe that, because he relied on pain relievers, he was a drug abuser. He was not. He was simply, like 100 million other Americans, a person in chronic pain. (2)
Primary pain conditions are among the most medically complex problems that providers face. (3) Compounding this complexity are misguided policy solutions as well as the historical, social, political, psychological, and legal realities particular to pain that leave a significant group of patients mistreated, undertreated, and untreated. (4) Decades of thoughtful, interdisciplinary policy work in the 1990s and early 2000s improved the environment for both the patients in pain and their providers." Some of that progress, however, is compromised by the understandable but ultimately incoherent responses to the reported rise in prescription drug overdose deaths in the United States (U.S.).
Several years ago, the Centers for Disease Control and Prevention (CDC) described increased rates of opioid related overdoses (OROs), (6) eventually declaring an epidemic of ORO deaths. (7) This triggered an unbalanced and disproportionate response by policymakers, practitioners, and the media that focused on opioids and their use in treating pain without a careful examination of the root causes. (8) Although the problems surrounding opioids and those surrounding the undertreatment of pain are empirically distinct, they are now conflated in ways that unnecessarily harm patients.
The epidemic of deaths associated with opioid use transitioned without notice into an "opioid epidemic;" (9) by 2012, Laxmaiah Manchikanti and colleagues omitted the words "overdoses" or "poisonings" altogether in publications and instead declared an "opioid epidemic with adverse consequences." (10) They blamed everything from liberalized laws, to the Joint Commission on Accreditation of Healthcare Organizations, pharmaceutical companies, (11) and even campaigns "touting the alleged undertreatment of pain." (12) The opioid epidemic language is now in mainstream use without qualification. (13) It is one of several common uses of infectious diseases vocabulary that attributes pathogen status to prescription opioids and patients in pain. (14) These responses come at the expense of patients already vulnerable to stigmatization and inappropriate treatment, and threaten to unravel decades of work to improve the regulatory and care environments. (15) Some providers and policymakers seem invested in the wholesale rejection of not only opioids, but also of patients themselves. (16)
Policy efforts focused almost exclusively on reducing the availability and use of prescription opioids without regard to predictable consequences of these use reduction strategies. (17) First, these strategies failed to acknowledge or address the specific nature of the OROs, most of which involved opioids combined with other substances and some of which were suicides. (18) Second, they failed to account for the serious underlying chronic and complex nature of substance use disorders, pain, and multiple comorbid conditions, such as suicidally. (19) Third, they proposed changes only to opioid use, not to the care of the patients involved. (20) Without considering the need for new approaches to treatment of patients with pain, substance use disorders, and the significant comorbidities, policy solutions left those suffering without assistance, and in some cases, increased barriers to treatment. (21)
Too many stories like that of Hyrum Neizer, a man with chronic, intense pain from debilitating headaches and suicidality, are symptomatic of the fragmented and distorted care many patients in pain receive. (22) Neizer's search for relief was characterized by unfruitful and humiliating trips to emergency departments (EDs) and doctors' offices. (23) He attempted suicide multiple times. (24) Although opioids relieved his pain, doctors stopped prescribing opioids altogether and convinced Neizer he was addicted to opioids. (25) He even admitted himself into group treatment for addiction, an experience he later described this way: "[M]y heart felt for them. But their stories weren't my story. I didn't have the desire to sell a kidney for drugs. I didn't want to rob pharmacies for OxyContin." (26) He only wanted his pain to stop. (27)
In 2016, Mr. Neizer may well have faced criminal prosecution under a doctor-shopping statute, (28) limited or no access to pain management physicians, (29) and providers even more reluctant or unable to prescribe opioids to treat his pain. (30) Myriad policy and practice level responses have emerged in the wake of a very serious public health problem of increased opioid abuse morbidity and mortality. (31) Many of these responses do not ad dress the actual harms--substance use related health impairment, functional decline, and premature death--and instead reflect a moral panic fueled by longstanding biases and stigmatization of individuals who have chronic or persistent pain (CP), (32) substance use disorders (SUD), (33) any mental illness (MI) or serious mental illness (SMI), (34) suicidally, (35) or a combination thereof.
This issue deserves attention and responses designed to address the complexity of issues surrounding pain treatment and SUDs. (36) Solutions must reduce overall harms without diminishing access and care to patients, including those in pain who benefit from opioid therapy. (37) Many of the responses to date lack the nuance required. Regulatory overreach, regression to old notions of patient legitimacy, and puritanical approaches to complex chronic conditions will ultimately do more harm than good. Instead, patient-centered laws and policies are needed, what Sandra H. Johnson has defined as those that serve the core values in medicine of relieving suffering, enhancing well-being, and increasing availability of effective treatments. (38)
This article will call for a careful examination of the facts, circumstances, and decision-making surrounding the recent round of restrictive policies regarding prescription opioids and the trickle down impact on the patients in pain and their providers. (39) Part II will provide an overview of the data and facts surrounding opioid-related injuries and deaths in context and what they mean for patients with pain and related conditions. Part III will explore the complexity of the problems of pain, and examine the related and co-morbid disorders that are often erroneously compartmentalized or ignored in practice and policy. Part IV will provide an overview of some common decision-making biases and errors and how these may be reflected in the responses of providers and policymakers to the opioid overdose epidemic. Part V will examine current reactions of law enforcement, legislatures, and policymakers, including legal frameworks surrounding the use of opioids in health care and their relationship to provider behavior. Part VI will endorse policy options synchronized to available data that do not devalue patients, or damage provider-patient relationships.
I. THE MISUSE OF OPIOIDS: REALITIES AND MINIMIZED CONCERNS
"If the competent and compassionate medical management of ... chronic pain ... were not already challenging, these have recently become more arduous because of the near hysteria that has attended the significant spike in prescription drug overdoses ." (40)
There are harms associated with drug abuse or misuse of any kind. (41) Premature death from drug poisoning (overdose) is the most serious of those harms, regardless of the drug(s) involved. (42) However, the responses to the problem of OROs are causing harm in their own right by 1) conflating the appropriate use of prescription opioids with a root cause of misuse and overdose, 2) labeling all overdoses as "opioid-related" no matter what role opioids play in the injury, 3) deflecting attention from other very serious trends reflected in these statistics, and 4) contributing to serious decision-making errors by policymakers and providers. (41) These collectively indicate that moralistic attitudes and ingrained sociocultural biases against people in pain and the use of opioids were only thinly cloaked by the serious advocacy efforts in the last twenty years. (44) Steven Passik described it this way: "People have returned to talking about opioids in religious terms, as if the drugs themselves are good or evil, and those emotions lead them to say, and even believe, things that are demonstrably false." (45)
Despite zealous policy reactions, very little progress to reduce the harms associated with misuse and overdose is evident, (46) while the increases in harm to patients living in pain are palpable. (47) In part, this is likely because the actual harms have not been carefully articulated; instead, prescription opioids and the patients in pain who benefit from them may be a stand-in for what is actually a network of public health problems with various degrees of overlap and intersection. (48) Slanted presentation of data has been described as an ethical problem in its own right; (49) this is because of the likelihood of harm to patients when policy and practice is based upon the "faulty mechanisms and procedures by which scientific data are interpreted for professionals, administrators, policy makers, news media, and the general public." (50) Synchronizing future solutions to actual harms requires a careful examination of existing facts.
A. Rises in Prescriptions and Opioid Related Injuries: An Illusory Correlation?
Opioid prescriptions have increased over the last three decades in the U.S., (51) as have the number of all prescriptions (52) These numbers are absent any context; the increase of opioid prescriptions may be, in part, a positive outcome of the efforts in the late 1990s to decrease physician fears of prescribing and improving the treatment of pain. (53) The rates of prescription opioid diversion and illicit use have also increased. (54) Most assume the rise in prescribing caused the increase in illicit use. (55) While there is a relationship between prescription rates and increase in illicit use, the available data does not support a direct doctor-to-patient-to-addict relationship; (56) instead, the correlation might be described as illusory, or "the tendency to perceive two events as causally related, when in fact the connection between them is coincidental." (57)
The majority of people who abuse, misuse, or overdose on prescription opioids are not the patients for whom they are prescribed. (58) The lack of relationship between receiving a prescription and misuse of the drug has been demonstrated by a variety of studies, including first-person accounts, as well as analyses of pharmacy and health care data. For example, a 2011 study by Cicero and colleagues showed the most prescription opioid abusers (over 86%) obtained opioids from dealers and from sharing with friends and family. (59) Only 13.8% obtained their prescription through their regular doctor or through doctor shopping (filling multiple prescriptions for opioids from multiple prescribes). (60) When the source of opioids was via a direct prescription, the prescribers were typically primary care providers rather than pain specialists. (61) The National Survey of Drug Use and Health (NSDUH) also indicates that the majority of those who misuse or abuse prescription opioids do so with pills that were not prescribed for them. (62) Further, McDonald and Carlson examined opioid prescription pharmacy data from 76% of retail pharmacies in the U.S. (63) Of the 146.1 million opioid prescriptions dispensed in 2008, only about 4% of medication dispensed was to doctor-shopping individuals, who constituted just 0.7% of purchasers; they purchased 1.9% of all opioid prescriptions. (64) In conclusion, "[t]he vast majority of opioid prescriptions involved a single prescription from [one] healthcare provider and most patients seemingly ... used them sparingly." (65)
Part of the reason there isn't a direct doctor to patient to addict relationship is that exposure to opioids alone does not create SUD. (66) Of course, some patients are more susceptible than others; factors such as a history of alcoholism or other substance abuse are more predictive of misuse of prescription opioids, but the overall rates of developing SUD after treatment for pain remain low. (67) A study by Cepeda and colleagues found that only three out of 1.000 people exposed to opioids go on to exhibit any doctor-shopping behavior (whether because of addiction or pseudoaddiction). (68) Those who receive an opioid prescription for an acute or temporary pain episode tend to take them as prescribed for pain and end up with leftover pills; literally thousands of tons of opioid prescription pills were collected in just a few years through drug take-back programs. (69) Thus, there is insufficient evidence that the careful use of opioids as prescribed for pain predisposes patients to future SUD or overdose. (70) Absent the predisposing genetic and environmental factors, "the drug does not have the power to change people in that way." (71)
Nonetheless, some groups and providers continue to respond as though prescription opioids are harmful in all circumstances. For example, Herzig and colleagues recently published an article in which they looked at the raw numbers of in-hospital opioid prescriptions. (72) They eliminated all patients with any type of surgical code, presuming they all had legitimate reasons for opioids. (73) They then analyzed the prescription rates for all non-surgical hospital patients with no additional context. (74) Considering no information such as diagnoses, histories, acuity of illness, or the like, the authors declare the prescribing practices inappropriate, (75) strongly implying doctors are placing their patients in jeopardy through improper medical practice. (76) A recent article by Kolodny and colleagues explicitly states without citation, "[t]he disease of opioid addiction arises from repeated exposure to opioids." (77) The authors also strongly recommend decreased pre scribing of opioids across the board, even suggesting that patients in pain should be deprived pain-relieving drugs to minimize the risk of diversion by family members or friends. (78)
B. Broad Definitions and Tunnel Vision in Nonmedical Use
The most cited statistics for the alleged epidemic of opioids are those that appear in survey reports from SAMHSA that track rates of self-reported non-medical use, emergency department (ED) visits, and deaths related to opioids. The primary source for non-medical use information is the annual NSDUH; however, information in the NSDUH on opioid use only dates back to 2002, the first year they started tracking opioids separately. (79) In addition, while the NSDUH survey asks about nonmedical use, the survey does not separate out whether a respondent took the prescription opioids to treat underlying pain or to "get high." (80) The NSDUH does not indicate any substantial changes in non-medical use of prescription opioids since 2002; (81) moreover, the 2013 and 2014 rates are lower than several of the previous years. (82) Furthermore, although all SUDs remain a serious public health problem, the overall rates of SUDs have been stable or declining over the last decade or more. (83) The rates of both SUDs and the non-medical use of opioids are stable over time; therefore, the increase rates of OROs are not attributable to a simple rise in misuse and abuse. In fact, the reasons for the increased rate of OROs are multifactorial.
1. Emergency Department (ED) Visits
The Drug Abuse Warning Network Survey (DAWN Survey) is the most common source of information about ED visits related to drug use and overdoses (poisonings). The DAWN Survey estimates alcohol, pharmaceutical, and illicit drug use-related visits to EDs based on surveys of selected metropolitan areas from 2004-to-2011. (84) The definitions used in the DAWN survey mean that many types prescription and illicit drug uses are counted as a drug related visit. For example, any one of the following conditions may qualify a visit as drug related: 1) the drug is part of the visit, whether or not it is the reason for the visit (85) or 2) any of the following criteria are met:
[a] taking more than the prescribed dose of a prescription drug; [b] taking more than the recommended dose of an over-the-counter pharmaceutical or supplement; [c] taking a drug prescribed for another individual: [d] taking a drug obtained illegally or without a legitimate prescription; and [e] deliberate poisoning with a pharmaceutical by another person; [or] [f] any use ... that ED medical staff document ... as misuse or abuse. (86)
Under this standard, a person who took four ibuprofen tablets over the counter, an acceptable prescription dose, or took their prescribed pain medication early because they were in pain--e.g., at three hours and fifty minutes instead of waiting the full four hours, or even people on a prescribed regime of Methadone, may be counted in these statistics. (87) While the survey authors explicitly acknowledge these limitations, many interpreters of the study have overlooked them.
The rate of visits to EDs for non-medical use of any pharmaceutical--any prescription, over the counter drugs, and supplements--increased by 132% between 2004 and 2011. (88) Visits for opioids increased by 183% over that same time period. (89) The increase is not just limited to opioids. For example, stimulants (drugs used to treat attention deficit disorders) increased 3 07%, (90) benzodiazepines (anti-anxiety agents such as Xanax or Valium) increased 149%,91 and anxiolytics, sedatives, and hypnotics increased 138%. (92) Moreover, between 2009 and 2011, there were no significant increases in opioid-related visits while other drug-related visits rose. (93)
The DAWN Survey report included important information about downward trends in the misuse of prescription opioids; however, it appears the misuse of other drugs may be filling the void. (94) Specifically, between 2009 and 2011, the rates of misuse of anti-anxiety and insomnia medications, stimulants, marijuana, and other illicit drugs increased while the overall rate of visits involving prescription opioids remained stable. (95) Coupled with the information about the stable rates of SUDs over time, it appears that individuals may be abusing drugs at the same rate while the drugs of choice are changing. As such, focusing on opioids alone rather than SUDs overall is a red herring. There are pressing needs for improved treatment efforts for individuals with SUD, education of patients about the proper use of and secure storage and disposal of medication, and the dangers of polysubstance use (mixing drugs and alcohol or mixing opioids and benzodiazepines).
2. Opioid Related Overdose (ORO) Deaths
"The reasons for the deaths are multifactorial, encompassing prescriber behaviors, patient contributory factors, non-medical use patterns, and systemic failures." (96) There are particular difficulties with the statistics on ORO deaths; for example, there are 1) no standardized definitions for post-mortem toxicology, 2) no standard qualifications or training for individuals who complete death certificates, 3) overlapping and confusing ICD-10 categories for death, and 4) no standard definition for "opioid-related death." (97) In addition, the spotlight on ORO deaths is somewhat disproportionate given very the numbers of other types of untimely deaths. For example, in 2010, all opioid-related deaths numbered 16,651 (only about 5,000 are attributable to opioids alone), while suicide killed 38,364 people. (98) Non-steroidal anti-inflammatory drugs, such as Motrin or Naproxen, are estimated to kill up to 10,000 people a year (double the number that opioids alone kill), (99) and approximately 88,000 deaths each year are attributable to excessive alcohol intake. (100) Why is the focus on the harms of opioids alone? Judy Foreman sums it up by saying "our collective thinking is out of whack." (101)
a. High Risk Polysubstance Use and Relative Opioid Risk
Oversimplification and broad-brush treatment has infected discourse around the dangers of opioids. The labeling of overdoses as opioid overdoses is one example. Opioids are rarely the only drug in the systems of individuals who experience OROs; in at least two-thirds of the cases, alcohol, benzodiazepines, or illicit drugs are also present. (102) This makes cause attribution difficult to impossible. Nonetheless, the default label is ORO, (103) however, labeling them "polysubstance related overdoses" seems more accurate and "benzodiazepine-related overdoses" is just as appropriate. The opioid-related label stuck and probably contributes to lingering one-dimensional concerns--as well as harms from ignoring the specific dangers of mixing prescription drugs and alcohol--through availability cascades. (104) Even though the dangers of polysubstance use were clear for over a decade, policymakers did not address the specific risks that mixing benzodiazepines with opioids or alcohol presents until 2014. (105) Earlier intervention aimed at these particular dangers most likely would have prevented harm and even some deaths.
Not all opioids are created equally--some opioids, like Methadone, are disproportionately risky. (106) Methadone accounts for less than 5% of all opioid prescriptions, but is involved in one-third of opioid related deaths, (107) as well as 30% of non-fatal overdoses. (108) The disproportionate danger relates, in part, to specific cardiovascular risks and physician knowledge deficits. (109) In fact, methadone use may be increasing. Between 1999 and 2006, Methadone-related ED visits rose sevenfold. (110) Several factors could account for increased use of Methadone: 1) it is often viewed as a less-addictive opioid alternative for pain treatment; 2) Methadone may be used as part of an opioid dependence plan (or medication assisted treatment (MAT)); (111) and 3) because it is less expensive than other opioids, many payer policies mandate its use. (112) Combined with the hazard of benzodiazepines and alcohol, these dangers deserve customized monitoring and interventions aimed at reducing overall harm, rather than the dominant blanket opioid use reduction efforts.
The broad-brush treatment pervades statistics as well. Inexplicably, the CDC groups suicides and homicides together with unintentional opioid overdoses. (113) This underlies the CDC's claim that opioid-related overdoses are the leading cause of injury-related deaths. Unintentional poisonings are actually the third leading cause of injury mortality behind 1) suicide and 2) motor vehicle accidents. (114) These are each serious problems that deserve attention; however, grouping together accidental and intentional overdoses obscures the root causes of the harm and obstructs synchronized harm reduction solutions.
Suicide is a growing public health problem in its own right and is now the leading cause of injury death in the U.S. (115) In 2010, one million people attempted suicide. (116) Suicide morbidity and mortality outnumbers opioid-related morbidity and mortality each year. (117) The rate of drug-related suicide attempts rose 41% between 2004 and 2011, (118) and 51% between 2007 and 2011. (119) In fact, since 2005, the number- and population adjusted rates of suicide have steadily increased from 10.9 per 100,000 persons in 2005 to 12.57 in 2013. (120) These premature intentional deaths deserve attention targeted at their underlying cause, especially because the very populations who have access to opioids have high rates of suicidality.
Addressing the underlying causes of morbidity and mortality in context is critical; yet, policy focuses on the existence of an opioid prescription absent context. Prescription opioids and the prescribers have become the "folk devils" in another moral panic in the "war on drugs." (121) When devising strategies of care and harm prevention, decision makers must understand the problem as complex, dynamic, and deserving of contextual solutions. Single-dimensional prescription opioid use reduction efforts are insufficient at best; at worst, they may foster the flourishing of continued suffering by ignoring the root causes of morbidity and mortality.
II. PAIN--ONE OF MULTIPLE INTERACTIVE, DYNAMIC, COMPLEX PUBLIC HEALTH PROBLEMS
"Pain and suffering--they go together like love and longing. Not the same thing, and not cause and effect, but so tightly woven that it's hard to imagine one without the other. (122)
The inappropriate treatment of pain is a "longstanding public health problem--some would say a public health crisis." (123) Pain alone affects at least 100 million U.S. adults (124) and is a "leading cause of disability and [a] major contributor to health care costs." (125) "The annual U.S. expenditures related to pain ... are higher than those for cancer, heart disease, and diabetes combined." (126) The total financial cost of pain to society ranges from $560-to-$635 billion, (127) far more than the $193 billion annual cost of all illicit drugs. (128)
The inappropriate treatment of pain is rooted in a web of entangled and relational barriers originating from systems, providers, and patients. (129) Systems-level barriers include formal legal and regulatory proscription, (130) as well as organizational policies and recommendations; at their core, individuals with varying levels of power, bias, and priorities influence them. (131) Disparities in pain treatment also reflect ingrained biases based on gender, (132) race, (133) socioeconomic status, (134) and other perceived differences. (135) Recent public health concerns surrounding opioid-related overdoses only magnify the complexity. Makota and colleagues found that providers' "fears surrounding opioids intersect powerfully with existing biases," (136) and resulted in disparate prescribing practices. (137) They also illuminate the inadequacy of a binary public health model of balance in which adequate treatment of pain is on one side and prevention of opioid misuse is on the other. The Pain & Policy Studies Group at the University of Wisconsin frames the issue this way: "There are important ongoing efforts in the U.S. to address simultaneously two major public health crises--1) the medical under-treatment of pain and 2) the non-medical use of controlled substances--both of which involve the opioid analgesic class of medications." (138) This is true but incomplete. Putting patients in pain in one box and those who might misuse opioids in another neither reflects reality nor provides the nuanced approach this complex problem requires.
Patients who request or take opioids are not either legitimately in pain or drug addicts. The picture is far more complicated. Those who misuse opioids are just as deserving of care and treatment as those who are on prescribed opioids that are functionally helpful. (139) Patients who request opioids may do so for a variety of reasons, none of which is mutually exclusive; most do so because they are in pain. Others may request opioids to feed an underlying SUD; those in CP may fear being without pain medication for future exacerbations and therefore hoard medication, while those with MI may use opioids to self-medicate. Any patient may hoard medication as part of a suicide plan. A small percentage of criminals (140) may deceive physicians into prescribing opioids for the purposes of diversion (as opposed to reasons related to underlying disorders). (141) Sometimes opioids are appropriate and other times they are not, for a multitude of reasons; context is fundamental to this assessment. Providers are obligated to carefully evaluate and treat each patient. Any of those patients may have one or more conditions that require additional attention and care, such as CP, MI, suicidality, or SUD.
Often ignored in scholarship and policy is that patients with CP, SUD, or MI are all equally deserving of respect and treatment; these are also not diagnoses of mutual exclusion. (142) They have much more in common than current policy recommendations reveal: (143) all are highly stigmatized, seriously undertreated, and grossly underfunded. (144) Providers regard individuals with these conditions as challenging, and those individuals express difficulty with access, discounting, and disbelief by providers. (145) Each diagnosis has high rates of comorbidity, such as suicidality, (146) that garner little attention in the literature. Thousands of pages are devoted to the harms associated with opioids and the screening of patients for illicit drug use, or for not using their prescribed opioids. Almost no literature draws attention to the very real and more widespread concerns about screening and treatment for SUD (as opposed to diversion), MI, and suicidality to prevent the serious associated harms, including premature death.
A mostly overlooked consequence of undertreated pain is the substantial risk of suicide. (147) The rate of comorbid suicidality in patients with CP ranges from 17%-to-66% of the population, (148) but even at the lowest estimates of 17%, it far exceeds the approximate 4% rate in the general adult population. (149) In fact, pain and disability perception (belief that one is disabled) are two important risk factors for suicide; (150) in patients with CP, disability perception coupled with a preference for death over disability is a significant predictor of suicidality. (151) Kanzler and colleagues found perceived burdensomeness was a strong predictor of suicidality in patients with CP, even suggesting the possible usefulness of a single-question screening tool for suicidal ideation. (152)
Individuals with certain types of CP syndromes may be at higher risk. (153) After controlling for comorbid psychiatric disorders, Ilgen and colleagues found a significant association between suicide death and three particular kinds of pain: 1) back pain; 2) migraine; and 3) psychogenic pain. (154) Many other studies have "demonstrated a link between chronic pain and suicidal ideation, planning and attempts." (155) Patients who have two or more painful conditions are also much more likely to attempt suicide. (156) First person accounts by patients with CP are infused with references to suicidality and the fact that for some patients, opioids are the only treatment that has prevented their suicide. (157) Yet, there are no current suicide prevention efforts that focus on pain as an independent risk factor. (158)
Approximately 90% of individuals who attempt or die by suicide have one or more Mis, whether or not they received a formal diagnosis or care for the MI. (159) Although many suicidal individuals did not receive mental health care in the year before death, the majority of them saw primary care and non-psychiatric specialty providers; (160) the same specialists likely to see patients with CP. MI is prevalent in patients with CP as well, with rates ranging from 30-60%.'61 Therefore, primary care and specialty providers are well suited, and arguably obligated, to assess patients for mental illness and suicidality and attempt to reduce the associated harms.
Patients with chronic pain are more likely to be suicidal than addicted, and yet, screening recommendations focus exclusively on detecting SUDs through complex risk stratifications and urine drug screenings. (162) The harms associated with suicidality are more serious than those associated with SUD; however, there are no practice recommendations or calls for universal suicide screenings of patients in pain beyond a handful of articles. (163) Further study of the relationship between pain and suicidality as well as urgent development of effective screening tools should be part of coherent practice and policy recommendations.
B. Mental Illness
Another area of concern in practice and policy is the failure to prioritize and address comorbid MI in patients with CP. Estimates of the extent of comorbid MI vary widely by population and treatment setting surveyed, although they exceed the rate of MI in the general population across the board. (164) For example, the mean prevalence of comorbid depression is estimated between 50% and 60% among patients treated in pain clinics and orthopedic and rheumatology clinics but only around 30% of patients in primary care clinics. (165) The rates of anxiety disorders are also substantially higher in patients with CP than in the general population. (166) Addressing comorbid MI in patients presenting with CP is delicate for providers, in part because patients often equate these concerns with being told their pain is imaginary. (167) Years of theory linking medically unexplained symptoms to somatization (i.e. emotional problems expressed through bodily ailments) worsen these concerns. (168) Cross training that prepares providers to care for patients with multiple complex problems is rare. (169) These comorbidities are often difficult to assess and diagnose both because of providers' lack of training and the patients' overlapping of symptoms of pain and depression. (170) It takes a skillful clinician and a thoughtful patient to sort out the problems, (171) all the while operating in an environment that rewards procedures over process and therapy. Howe & Sullivan have advocated that every patient presenting in CP should receive a comprehensive assessment of psychological health and appropriate referrals when needed. (172)
Lack of access to or payment for mental health care and integrated multidisciplinary care are significant obstacles to appropriate care. The communication and knowledge barriers to comprehensive assessment compound the problems, leaving many patients with undiagnosed or untreated comorbid MI and, possibly, suicidality; (173) this only worsens morbidity and mortality. Harm reduction strategies in the future should include screening patients for depression and other conditions and appropriate referrals if needed. (174)
C. Substance Use Disorders (SUD)
"[P]atients with substance use disorders and pain have the right to be treated with dignity, respect, and the same quality of pain assessment and management as all other patients. " (175)
Acknowledging that patients with CP may also have SUD, or even engaging in a dialogue about potential for abuse, is even more delicate than addressing mental illness. In addition to stigmatization, the behaviors associated with SUD are more likely to lead to criminalization than treatment. (176) Moreover, patients in pain tell a near universal story of at some point having their reported level of pain or motivations questioned by providers, with providers suspecting them of drug seeking, or criminal activities associated with drug diversion. (177)
The inclusion of opioids as one tool in the toolbox of pain treatment occupies a unique position in medicine. Physicians often prescribe drugs that are also misused or abused for conditions other than CP and physicians often care for patients engaged in illegal activity. Only in the realm of prescription opioids have providers' obligations transformed from care to criminal investigation. (178) Until law and policy allow providers to embrace the complexity of pain coupled with an obligation to provide care and treatment or referral to patients with SUD, these interactions will not improve. (179) This will require a reexamination of the kinds of prescribing that warrant scrutiny and the expectations of providers and policy makers.
Drug policy in the U.S. has been predominately one of shock and awe, focused nearly exclusively on what Reuter and MacCoun call "use reduction." (180) Drug use is easier to measure than levels of hann--and only measuring the use of prescription opioids is even easier--but it does not get to the heart of the problems. Use reduction is grounded in moralistic, criminal justice approaches, rather than a public health approach of harm reduction. (181) According to Mark Kleiman,
It turns out to be substantially easier to announce that one is opposed to drug-taking than to craft public policies to reduce the damage it does. ... A policy of announced hostility toward drug taking and drug-takers will tend to make the remaining drug takers worse off, and more dangerous to others, than they would have been otherwise. Moreover, it leads citizens and their representatives to shy away from the part of drug abuse control policy that involves providing services to drug-takers to help them quit, moderate their behavior, or better integrate themselves into the broader society. (182)
The rejection of a harm reduction model has trickled down into poor access to care and a lack of providers for SUD. (183) Indeed the common construction of patients with "real" pain as legitimate expressly excludes the possibility of comorbid SUD, leaving individuals with addiction portrayed as outgroup members, illegitimate, and undeserving. (184) Changing the approach to address SUD, especially discussing it with patients in pain, will require policy makers and providers to address their own underlying biases and to embrace the goal of harm reduction for their patients. (185)
Current research suggests the rates of comorbid CP and SUD are variable. (186) Within this discussion, there are two distinct ways of viewing the degree of crossover: 1) those with a primary diagnosis of SUD who also report significant pain, and 2) those with a primary diagnosis of pain who develop an opioid use disorder during treatment. (187) The numbers for the first are well above the rates of pain in the general population. (188) The numbers of people with CP who develop a SUD without a history of substance abuse are around the level of SUDs in general population. (189) Providers tend to overestimate the rate of the second group, leading to avoidance of opioids in otherwise appropriate cases and undertreated pain in many patients (with or without SUD). (190) Even among patients with a history of substance abuse, physicians are reluctant to address the risks in the context of pain treatment, leading to undertreatment. (191) Work by Joseph Merrill and colleagues indicates providers have great difficulty navigating the comorbid treatment of pain and SUD and that patients may suffer as a result. (192) They found several common themes including providers' fears of being deceived, use of non-standardized approaches to pain assessment in patients with SUD, and avoidance of patients; patients expressed fears of mistreatment or punishment by providers for their SUD. (193) An integrated care approach for patients with pain and comorbid conditions, such as the one explained by scholars Teresa Jacobson and Gregory Hatchett, is needed to emphasize holistic, contextual assessment, and ongoing care. (194) An integrated approach would do far more to reduce morbidity and mortality of these complex chronic conditions than narrow, opioid focused supply side efforts. (195)
D. Providers' Obligations to the Patient & Self-regulation Through Virtue
"Physicians require the virtue of humility (understood as self-knowledge and an openness to the perspective of others rather than as meekness) to support use of the habits, or 'compensatory strategies,' that will enable physicians to prioritize the goals of medicine over their own self-interest. " (196)
There is certainly evidence that many providers are able to work through the complexities and competing concerns involved in treating pain, including prescribing opioids. (197) Patients would rather be heard and deemed trustworthy than receive any particular therapy or treatment. (198) They repeatedly express their desire to be listened to and legitimized. (199) Bergman and colleagues recently identified frustration among CP patients who felt providers were disengaged, finding many patients "wanted more priority placed on discussing pain ... expressed desire for their pain to be recognized as real, [and] were not seeking to discuss opioids; many were in fact avoiding opioids and looking for a sympathetic ear." (200) They are perhaps looking for communication that incorporates elements of Epstein and Gramling's work on collaborative cognition, which they describe this way,
Engaging patients in constructing preferences in the face of complexity, inadequate evidence, and irreducible uncertainty involves more than provision of information and an invitation to choice ... [it] is relational, dynamic, iterative, provisional, and conditional--it involves building relationships, providing information, and exploring preferences, which then strengthen relationships, understanding, and involvement in decisions. (201)
The importance of humility and attending to the patient in context and considering possible factors that may unduly influence decisions is advanced by many experts under many names, ranging from mindfulness to self-monitoring to empathy to cognitive debiasing. (202) These strategies are effective for appropriate care. (203) The IOM also called upon providers to develop care that is patient-centered, comprehensive, and interdisciplinary. (204)
Individualized, contextual practices reflect providers' adherence to certain dispositions that advance the ends of medicine and enable providers to prioritize well-being in the context of the patient's particular vulnerabilities and needs. (205) In order to achieve those ends, especially in this context, virtue is paramount. DuBois and colleagues explained virtue this way: "[Dispositions (or traits, in the language of psychology) define how we behave when no one else is watching; accordingly, they serve as a bedrock for professional self-regulation, particularly at the level of the individual physician." (206) In the specific context of pain medicine, James Giordano has argued for what he calls "an agent-based virtue ethics of pain medicine;" (207) a necessity because of the unique and profound character of pain and the need for both equanimity and empathy in caring for the patient. (208) Combining the work of Dubois and colleagues with Giordano's creates a virtuebased ethic in the care of patients in pain that emphasizes humility (including self-knowledge, reflection, and intellectual honesty), compassion, empathy, and practical wisdom. Embracing a virtue-based ethic also allows providers and policy makers to more readily avoid common decision-making errors that adversely impact patients.
III. DECISION-MAKING ERRORS BY PROVIDERS & POLICYMAKERS
It is hard to understand why decision makers ranging from individual providers to policymaking bodies craft solutions that are inconsistent with the problems they are meant to address. In terms of policymaking, researchers have discovered that factors such as public opinion and salience (the degree to which the issue stands out against others) are significant influences. (209) Other research indicates that policy makers, at best, only indirectly use public health research and evidence to inform policy recommendations; (210) therefore, it is not surprising that policy solutions are not synchronized to problems. There is some evidence that a better understanding about how and why providers and policy makers make decisions can improve their decisions in the future. (211)
A. Social Cognitive & Moral Decision-making
"A more refined appreciation of human tendencies--both their operation and their possible origin--may help us to better understand what educational and policy interventions may facilitate good conduct and ameliorate bad conduct. " (212)
Ultimately, both policy and provider level decisions about opioids are in the realm of ethics and moral decision-making, (213) and strongly impact patients with pain or SUD. (214) James Rest's interdisciplinary model of moral decision-making includes four distinct but dynamic, non-linear components: (215) 1) awareness or sensitivity (capacity to recognize that a situation has moral content); 2) judgment (evaluation and reasoning between options and attendant consequences); (216) 3) intention or motivation (commitment to choose one of the options that is the most morally right, even in the presence of choices that offer more personal gain); and 4) action (enacting the choice). (217) Failure of one will weaken or may prevent a person from making the ethically appropriate choice. (218)
Moral decision-making is also influenced by the degree of proximity, social consensus, and the magnitude and extent of the consequences. (219) Proximity influences moral awareness. (220) Low levels of empathy for patients in pain may obscure moral awareness. (221) This concept complements other research describing the persistent stigmatization and discounting of patients in pain (including those with MI and SUD), (222) as well as my previous work on providers' moral disengagement of patients in pain, particularly when providers do so by sullying the victim (or patient). (223)
The strongest contributors to moral decision-making overall are social consensus, the magnitude of the consequences, and the probability of effect. (224) In particular, social consensus--expressed informally through group interactions and practices or formally through rules or laws--is a powerful mediator of moral awareness, judgment, and intention. (225) Thus, in the context of decisions about patients in pain, the attitudes of other providers as well as organizational policies and regulation may heavily influence decisions. (226) In addition, the extent and likelihood of the perceived consequences of a decision maker's actions may influence their willingness to act. (227) For example, if providers fear regulatory scrutiny or even prosecution for prescribing opioids in an uncertain clinical encounter, they are less likely to consider a prescription a viable option. (228) Likewise, if they are (reasonably or unreasonably) concerned about addiction, the less likely they are to consider an opioid prescription. (229) The degree of these beliefs, with or without supporting evidence, may contribute to incoherent decision-making. (230)
Decision-making is part of social cognition. (231) A wealth of literature exists on theories of decision-making in general, (232) in health law & policy, (233) and in the context of medical diagnosis. (234) Unfortunately, very little research exists on decision-making errors in ongoing treatment decisions (as opposed to diagnosis), (235) in the context of pain, (236) or on the policy decisions surrounding controlled substances, including opioids, and patient care. This is critical because clinical encounters with patients in CP are highly complex, objective findings are often inconclusive, and options are ambiguous. (237) Acknowledging the moral and clinical complexity of caring for patients in pain may allow providers to adjust their approach to decision making and patient engagement. (238)
B. Dual Process Models of Decision-making
Decision-making involves complex coordinated neurological computations and processes and decision-making strategies often change through experience. (239) Dual process theories dominate discussions of decision-making, (240) as well as theories surrounding the existence of moral intuitions and their relative value. (241) In the context of medical decisionmaking, Croskerry and colleagues offer a thoughtful overview of psychological factors that influence cognitive performance in provider decision making through a discussion of the Dual Process Theory (DPT) and the influence of cognitive biases and distortions on decisions. (242)
Under DPT, decisions are made in either an intuitive (Type One) or an analytical (Type Two) mode; (243) "clinical expertise thus depends on the ability to move back and forth between the two modes." (244) The default mode, Type One, is unconscious, fast, and efficient; (245) it is also "characterised by heuristics--short-cuts, abbreviated ways of thinking, maxims, 'seen this many times before', ways of thinking." (246) Heuristic use in Type One processing is a function of attribute substitution, ignoring "part of the information, with the goal of making decisions more quickly, frugally, and/or accurately than more complex methods." (247) Heuristics function efficiently, by filling in missing information or by saving the mental work of analyzing a large amount of information. (248) Although always efficient, they are not always effective. (249) An understanding of the difference is crucial to appropriate decision-making. (250)
The fuzzy-trace model expands upon the DPT by describing two distinct forms of Type One decision-making: "[impressionistic thinking using vague gist representations" (impressionistic thinking) and "insightful intuition." (251) Insightful intuition includes knowledge assimilation--such as emergency treatment protocols that through learning and practice transition from Type Two into Type One. (252) Type One may also be responsible for the gut feeling or sense many clinicians describe that can contribute to good care. (253) On the other hand, impressionistic thinking may include the culturally assimilated bias against patients who have a SUD. (254) Thus, the use of Type One processing is neither inherently good nor bad. While Type One processing is indispensable in certain situations, such as lifesaving procedures, other situations call for more reflective reasoning. (255)
Type Two processing is the reflective, analytical mode involving metacognition (thinking about thinking); it is "fairly reliable, safe and effective, but slow and resource intensive." (256) Type Two processing occurs whenever the actor is consciously thinking about what she is doing, such as following a checklist, weighing options, or questioning his or her own as sumptions. (257) Using the wrong rules of analysis contribute to decision-making errors in Type Two processing as do factors that contribute to cognitive overload. (258) Errors related to bias are more common in Type One mode; Type Two processing is frequently required for correction. (259)
C. Biases & Heuristic Failure in Decision-Making Errors
Certain situations compound the risk for biased reasoning or failed heuristics, what Wilson and Brekke call "mental contamination." (260) Some of these situations are characteristic of providers' encounters with patients with CP, as well as those with comorbid MI or SUD. According to Graber and colleagues, "[e]rrors are more likely when the level of uncertainty is high, if clinicians are unfamiliar with the patient, and when there are atypical or non-specific presentations ... or distracting comorbid conditions." (261) Decisional errors are common when they involve individuals in groups prone to stereotyping, such as patients with CP, SUD, or MI. (262) According to Klein, "the greatest obstacle to making correct decisions is seldom insufficient time but distortions and biases in the way information is gathered and assimilated." (263) While no researcher has specifically examined providers' and policy makers' cognitive biases and failed heuristic in the context of patients with CP, an examination of some of the potential heuristic errors and biases is warranted.
1. Confirmation and Anchoring Biases, Representative Heuristic
One of the most well described biases is confirmation bias, (264) a "tendency to look for, notice, and remember information that fits with our preexisting expectations ... information that contradicts those expectations may be ... dismissed as unimportant." (265) It is a way to resolve cognitive dissonance (the discomfort of holding two contradictory ideas simultaneously) by interpreting subsequent information to fit the initial idea. (266) Epstein and colleagues describe the interaction between Type One processes and confirmation bias as a consequence of biology: "[O]ur brains--evolved to guess the most plausible interpretation of the limited evidence available, in which the mind 'imposes a definition on things and then mistakes the definition for the actual experience'--and also ignores disconfirming data." (267) The operation of confirmation bias can jeopardize accurate and appropriate treatment. (268) Anchoring bias is related; it occurs when an incorrect initial impression is made and then all subsequent work focuses on that incorrect impression. (269) Ely and colleagues describe it as "The tendency to perceptually lock on to salient features of the patient's presentation too early ... and failing to adjust this impression in light of later information." (270) Rather than just selectively interpreting subsequent evidence to fit the initial impression, as is the case with confirmation bias, decision makers focus all efforts on the initial idea. This, of course, can work in concert with confirmation bias to lead decision makers to both attend more heavily to information that confinns their assumptions and proceed to make subsequent decisions based on the initial anchor.
An overlapping problem is reliance on the representativeness heuristic; (271) this is the tendency to look for prototypical manifestations while failing to consider atypical presentations. (272) Decision makers may misjudge the actual situation and ascribe more value to one piece of information. (273) This is particularly concerning when a large variety of patients are grouped together, as is the case with chronic pain patients or patients with malignant (cancer) pain verses non-cancer pain, a dubious distinction at best with no "physiological, pharmacological, or even philosophical  basis." (274) This distinction dates back to a time when cancer pain really meant pain at the end of life. (275) The distinction is now without meaning. For example, does the appropriate treatment of post amputation pain change whether the amputation was caused by a tumor or a traumatic injury? (276) Others have advanced compelling examples to illustrate the absurdity of the distinction, such as the exclusion of patients with sickle cell disease (a painful, chronic non-cancer condition), (277) or patients with painful, terminal non-cancer con ditions. (278) These kinds of distinctions also illustrate the representative heuristic that may operate in decision makers' reliance on disease as an entity independent from the patient. (279)
The negative impact of the operation of these biases may account for the predominant notion that providers can tell upon first impression whether a patient is actually in pain or "on the level." (280) Perhaps a patient appears comfortable and happy when the provider sees him or her sitting in the exam room; when he or she reports severe, even crippling pain after walking to the exam room, the provider may decide he or she is not "on the level" and ignore other information. (281) In reality, this is a typical presentation with lumbar stenosis. (282)
Confirmation and anchoring bias may combine to explain the unfortunate effects of provider, institutional, and regulatory reliance on red flags for diversion. (283) One such red flag is when a patient asks for a particular opioid drug by name. Suppose a patient said, "Vicodin makes me itch, but Percocet worked well for me when I hurt my back a few years ago." (284) If the provider has decided the patient is not diverting based on an initial impression, they may interpret the patient's statement as an indicator she is a good historian. Otherwise, she would quickly be suspected of diverting and denied a prescription that would otherwise relieve the acute pain.
Confirmation and anchoring biases in this arena are not limited to providers. For example, the Inspector General of Tennessee believes that providers can intuitively know the difference between a patient in pain and someone wishing to divert, saying "[i]t's not easy for a physician or pharmacist to be able to tell the difference between a legitimate patient and a drug abuser, but providers in Tennessee have developed a good sense of distinguishing the abusers." (285) Dr. Gary Jay expressed dismay at the negative experiences of some of his long-time patients after Walgreens, a national drug chain, adopted the use of red flags to deny filling prescriptions for opioids. (286) One red flag (or anchor) is enough, without other contextual information, to justify denying the patient a prescription. (287) Payment in cash, one red flag, is grounds for denying patients their prescription, regardless of the reasons. (288) Another red flag is multiple pharmacy customers with the same diagnoses and prescriptions from one provider. (289) Other specialty physicians, such as pulmonologists who have multiple patients with asthma all prescribed bronchodilators, are not subject to the same suspicion as similarly situated pain physicians.
2. Availability Bias and Availability Cascades
Availability bias (or availability heuristic) occurs when the likelihood of an issue is "tied to the ease with which its occurrence can be brought to mind." (290) Inordinate weight is placed on examples or categories of previous situations, with stronger memories often afforded more credence. (291) The corollary to availability bias is base-rate neglect, which is the tendency to ignore the true prevalence of disease. (292) Combined, they "can lead to serious errors of fact, in the form of excessive fear of small risks and neglect of large ones." (293) The media, public health, and scholarly attention to OROs make the risks of opioid misuse more available than concerns with objectively greater morbidity and mortality. (294) This may explain why providers and policy makers overestimate risks of addiction, discount the extent and consequences of persistent pain, and neglect assessments for MI and suicidality. (295) Availability bias may also extend to the decisions made by coroners and physicians in selecting a cause of death on death certificates. The significant publicity around opioid related deaths may increase the attribution of death to opioid poisoning rather than one of the multiple other drugs or alcohol present in the systems of most victims. (296)
Availability bias may explain some policy decisions; for example, the disproportionate focus on opioids over polysubstance abuse concerns may be a product of availability bias. Likewise, the significant attention to diversion detection in CP treatment guidelines with little to no attention to greater causes of morbidity and mortality may arise, in part, from availability bias.
Poor opioid policy and provider decision may be a product of availability cascades, a phenomenon that Kuran and Sunstein explain as an interaction between initial availability heuristic and social mechanisms resulting in snowballing or bandwagon effects of "persistent social availability errors." (297) The social mechanisms of informational and reputational cascades are part of availability cascades. (298) Information cascades occur when individuals with incomplete information rely on others (often too with incomplete information) in formulating beliefs. (299) Reputational cascades are motivated by social approval needs and occur when individuals choose to indicate, or refrain from rejecting, the beliefs of others. (300) These interdependent cascades can snowball and become self-reinforcing for the groups that espouse particular views and "public discourse will rest on flawed judgments," (301) leading to serious social harm through narrow and ill-informed policy making, media attention, and a focus on slight risks at the expense of important risks. (302) Availability cascades "constitute a major ... source of the risk-related scares that have cramped federal regulatory policy at both the legislative and executive levels, with high costs in terms of lives lost, lowered quality of life, and dollars wasted. ... [C]ascades force governments to adopt expensive measures without careful consideration of the facts." (303)
Some of the recent regulatory efforts aimed at reducing opioid use may be incoherent because of availability cascades. As initial policy makers and public health agencies focused on prescription opioids and pain practitioners alone, others adopted their views without a careful examination of the facts. As time went on, reputational fears may minimize and silence dissenting views, resulting in a one-dimensional, prescription opioid use-reduction approach to the problem. For example, state guidelines that ultimately decrease access to opioids for all patients come with significant administrative and enforcement costs, as well as social costs such as patient suffering and even increased use of more dangerous illicit drugs like heroin. States laws that mandate urine testing for all patients taking opioids create significant financial costs to individual patients and third party payers (304) Prescription drug monitoring programs that are costly but do not report in real time are almost useless in detecting diversion. (305) These policies that are not synchronized to the problems they aim to address, discussed in Part V below, may be products of availability cascades.
3. Framing Errors and Illusory Correlations
The ways in which antecedent conditions of decisions are framed strongly influences subsequent decision-making. (306) For example, the same surgery described two different ways elicits different decisions; more people agree to the surgery when told "ninety percent of people are alive after five years" than when they are told "ten percent are dead after five years." (30) It is not surprising that when reports frame opioids as deadly and "causing" overdoses--with little or no mention of the therapuetic utility for some patients in pain--that people are willing to sacrifice pain control for a perceived greater good. Framing polysubstance deaths as OROs instead of deaths from mixing alcohol and multiple classes of drugs only furthers the scapegoating of opioids alone. Most concerning, this may have delayed or decreased critical warnings about the risks of polysubstance use and contributed to untimely deaths. It may also cause wholesale avoidance of the entire class of drugs by providers and patients, even when they may be appropriate and helpful in some circumstances.
Another impairment to decision-making is narrow framing, or the "tendency to define our choices too narrowly, to see them in binary terms." (308) The false binary pervades this area. Treatments are seen as bad (dangerous, addictive, and deadly) or good, with opioids falling in the "all bad" category. Patients are seen as legitimate or illegitimate, a pain patient or a drug-addict, a complete manipulator or a "straight shooter," deserving or undeserving. Why a patient with SUD or other related conditions is seen as illegitimate or undeserving is hard to understand. (109) Pain is also grouped into nonsensical categories, such as cancer and non-cancer pain or surgical and non-surgical pain. (310) Finally, the narrow framing and interpretation of the principle of balance in prescribing is illustrative. (311)
4. Visceral Biases: Indignation, Outrage, and Betrayal
Personal feelings and emotions are powerful drivers of poor decision-making. (312) Sunstein describes particularly strong feelings as heuristics in their own right; (313) the betrayal of trust heuristic is so strong, according to Sunstein, that people will actually substantially increase risks to themselves simply to avoid betrayals of trust. (314) Closely related are the feelings of outrage and indignation, both of which combine with betrayal to create disproportionate perceptions of threat and desire to act to reduce the threat at any cost. (315) Worse yet, when groups of people prone to share indignation and betrayal deliberate together, the groups "end up more indignant than their median member." (316) Although not studied in the context of health care generally or pain specifically, this has serious implications for providers caring for patients who present in pain. Nowhere in the practice of medicine are the feelings of betrayal and indignation stronger than around the idea of a patient deceiving a prescriber to obtain opioid for diversion purposes. Matoka and True described providers' reactions this way,
[W]e were struck by how merely noting a research interest in pain (in the most general of terms) often elicited powerfully charged emotion, prompting clinicians to offer up expressions of frustration, anger, and even disgust in vivid terms'. "Ugh, pain patients--I hate those back pain guys. I just want to turn and run when I see one coming." And "Pain patients, well, you've picked a doozy there. What a waste, the kind of energy they spend trying to get their meds--makes me sick!" Notably, these sorts of virulent--even visceral--reactions often existed right at the surface ... clinicians frequently seemed so willing to talk about their patients with pain precisely because it is an area of their daily practice about which they often feel a deep sense of vulnerability, unease, and even failure. (317)
If the betrayal and indignation heuristics operate the same way in this context, it could explain why providers believe far more patients are addicted to or are attempting to divert opioids than the evidence supports. It could also explain their willingness to risk unnecessary patient suffering in favor of avoiding any personal risk of betrayal. According to Sandra H. Johnson,
What the debate between deceived doctors and earnest pain management advocates also often misses is the emotional burden that deception exerts on physicians and the behaviors that those emotion-laden circumstances produce. Absent recognition of the emotional state of mind of physicians in practice, however, it is unlikely that persistent calls for more trust between patient and physician will achieve the desired outcome. (318)
5. The Opioid Heuristic and Provider and Institutional Practices
"If treating certain conditions increases the risk of being called a bad doctor, many doctors will focus their efforts elsewhere. Doctors are, after all, only human," (319)
It is no surprise doctors are avoiding patients in pain. Now that the phrase "opioid epidemic" has pervaded the professional literature and public discourse, the landscape for patients in pain on therapeutic opioids is bleak. The epidemic metaphor feeds and intensifies availability cascades around prescription opioids; opioids are no longer value neutral tools with positive and negative effects. Value neutral tools are not equated with contagion and plagues: these are things to be avoided at all costs. (320) There is now what I call an opioid heuristic; opioids are standing in for a host of possible negative sequelae and triggering avoidance by providers. Quite literally, the word opioid has replaced "opioid related overdoses," ascribing the power of danger on contact for any reason. This is related to the older concept of opiophobia (fear of opioids); however, the trajectory of opioid-related public discourse has transformed the fear of opioids into a short cut in decision-making. The opioid heuristic is consistently reinforced by the epidemic metaphor as well as the opioid related availability cascades.
For decades, providers have contended that they fear the use of opioids because of potential investigations by the DEA or state boards for prescribing violations. (321) This fear is understandable given it is one of the few areas in which prescribers could find themselves embroiled in federal and state criminal, administrative, and even civil proceedings. (322) Others fear causing addiction in their patients, (323) a fear likely fueled by resurgence of the word "addiction" related to use of opioids in the last five years, in everything from media sources to statements of some provider groups. (324) While concerns about diversion and addiction are not unreasonable, they are disproportionate in comparison to other risks and should not obscure the comprehensive evaluation and treatment of patients. (325) Most importantly, these concerns must not eclipse the overall treatment of a person in pain, with or without the use of opioids; opioids are just one tool in the diverse and context-dependent treatment of patients in pain. (326)
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Introduction through III. Decision-Making Errors by Providers & Policymakers, p. 1-46|
|Author:||Dineen, Kelly K.|
|Publication:||Law and Psychology Review|
|Date:||Jan 1, 2016|
|Previous Article:||Psycho lawyer, qu'est-ce que c'est: the high incidence of psychopaths in the legal profession and why they thrive.|
|Next Article:||Addressing prescription opioid abuse concerns in context: synchronizing policy solutions to multiple complex public health problems.|