Pain medication narratives : exploring attitudes towards pain and risk perception of opioid painkillers among Americans in the U.S. and American expats in the Netherlands

(1)

University of Amsterdam, September 2017 - 2018

Master’s Thesis: Sociology, Social Problems & Social Policy (MSc) Submission date: August 1, 2018

Supervisors: Dr. Christian Bröer (first reader) & Dr. Patrick Brown (second reader)

(2)

Pain Medication Narratives

Exploring Attitudes Towards Pain and Risk Perception of Opioid

Painkillers Among Americans in the U.S. and American Expats in the

(3)

1.0 INTRODUCTION

1.1 Contextual Relevance & Existing Research

Over the past decade, drug overdose has become the leading cause of accidental death in the United States, killing 63,000 Americans in 2016. Approximately two-thirds of these deaths were opioid-related, involving both illicit drugs such as heroin and prescription drugs like OxyContin (CDC, 2017b). The significant role of pharmaceuticals in what has been deemed “the opioid epidemic” is of particular interest, as almost half of all opioid-related deaths involve prescription opioids (CDC, 2016b). Prescription opioids are a class of “painkiller” drugs that are used to reduce the sensation of pain by combining with brain receptors, most commonly known under the pharmaceutical brand names of OxyContin, Percocet, and Vicodin.

While effective for relieving pain, prescription opioids are highly addictive and can lead to instances of misuse, i.e., using the medicine in a way or dose other than prescribed, using someone else's prescription medicine and/or using the medicine to get high (National Institute of Drug Abuse, 2018). Instances of misuse can be fatal since opioids also attach to receptors in areas of the brain that are responsible for respiratory control, which can decrease the flow of oxygen to the point in which a person stops breathing. In 2016 alone, over 2 million Americans misused prescription opioids, resulting in approximately 16,800 deaths (SAMHSA, 2017). It's been argued that widespread prevalence of pharmaceuticals in American society guarantees that individuals will have some degree of experience with pain medication, whether it's using them first hand or witnessing them being used by their peers (Carter & Winseman, 2003; Quintero et al., 2006). Recent research highlights the magnitude of the problem concerning the unpreceded

(6)

related deaths. In 2016 alone, an estimated 2 million people suffered from OUD, with

approximately 90 Americans dying from an opioid-related drug overdose every day (SAMHSA, 2017; Bonnie et al., 2017). Furthermore, a recent survey conducted by The National Safety Council (2017) found that 1 out of 10 Americans knows someone who has died from an opioid overdose.

In response to the growing problem, the U.S. federal government declared a public health emergency in 2017 to address the opioid epidemic, which has since fueled subsequent action from public health agencies to devise a plan that aims to tackle the problem by “reducing the burden of suffering from pain while containing the rising toll of the harms that can result from the use of opioid medications” (Bonnie et al., 2017, p. 1). Among these efforts have been new guidelines from the Centers of Disease Control (CDC) that recommend that over-the-counter pain medication such as ibuprofen should be the first-line treatment and the typical protocol for prescription opioids should not exceed a three day supply (CDC, 2016b).

While implementing more restrictive prescribing guidelines has aimed to reduce the number of opioid-related harms, it has also raised concerns over whether it’s preventing patients from receiving treatment they have long relied on and need, particularly those suffering from chronic pain, which is defined as lasting for more than 3 months (Bonnie et al., 2007). Furthermore, there has been speculation that stricter prescribing laws have contributed to unintended consequences such the rise in heroin-related deaths, through diversion from

prescription to illicit opioids (CASA, 2005), where approximately three out of four illicit opioid users report abuse of prescription painkillers prior to using heroin (CDC, 2017). ). It’s widely believed that prescription opioid misuse is a risk factor for heroin use, as stated “parallel to an increase in prescription opioid overdose, heroin overdose deaths are increasing as patients shift

(7)

to cheaper and more accessible heroin from the prescription opioids which physicians are

prescribing less often due to increasing regulatory restrictions and increased awareness about the risk of addiction and abuse of the pain drug” (Smith, 2017, p. 95). Research on the relationship between prescription opioids and heroin suggests that prescription opioid abuse is a significant risk factor for heroin use, as the drugs are chemically similar (Rudd et al. 2016; Compton et al., 2016). In this regard, the opioid epidemic has demonstrated itself to be a multifaceted problem that lies at the intersection of pain, medicine, and drugs in American society, where “the complexity of pain is matched by the complexity of achieving the appropriate use of opioids” (Bonnie, Ford & Phillips, 2017, p. 3).

Prior to the 1800s pain was believed to be an “existential experience” that resulted from aging, but with the turn of twentieth century a new era of medicalization began taking shape in regard to pain medication (Tompkins, Hobelmann, Compton & Bonica, 2016).The 1990s has been considered a transformative decade in the making of today’s current opioid crisis, following a sequence of causal factors that reshaped the conceptualization of pain and practice of medicine in American culture (CASA, 2005). Among these developments included new guidelines from the American Pain Association (APA) that recognized pain as a measurable health condition and pushed for a change in philosophy around the use of prescription opioids for first-line treatment for pain. Additionally, the expansion in insurance coverage of prescription medication alongside aggressive pharmaceutical marketing efforts promoting opioids for chronic pain treatment. The United States is one of only two countries in the world, along with New Zealand, where direct-to-consumer (DTC) advertising of pharmaceuticals is legal, i.e., involving the promotion of prescription drugs through mainstream media channels (Ventola, 2011). The Food and Drug Administration (FDA) Modernization Act of 1997 prompted increased media promotion of

(8)

medical products and services with new regulations, as "pharmaceutical industry spending on television advertising increased six-fold between 1996 and 2000" (Conrad, 2005, p. 5). Among these promotional efforts was OxyContin's introduction on the market in 1996 was followed by a spike in sales from $48 million to almost $1.1 billion in 2000, and nearly a decade later

OxyContin was the leading drug of abuse in the United States (Van Zee, 2009). Between 1992 and 2016, the number of opioid prescriptions dispensed by doctors steadily increased from 112 million to 236 million (Pezella et al., 2017). In 2016 alone, 61 million Americans (19% of the population) received one or more opioid prescriptions, with the average patient receiving 3.5 prescriptions (National Safety Council, 2017; CDC, 2017). Opioids continue to be

overprescribed in terms of higher doses than needed and for more days than needed, which increases the risk of addiction and abuse potential (National Safety Council, 2017; Pollini et al., 2011). DTC advertising has been argued to have played a role in the widespread misconception regarding the addictive nature of prescription opioids coinciding with unprecedented rates of opioid prescribing and subsequent incidences of addiction, abuse, and overdose (NIDA, 2017). The intersection of pain, medicine and drugs in American culture becomes problematic when considering the notion that there is a misconception that prescription opioids are safe since they are provided by doctors and used for medical purposes (NIDA, 2016), without considering the risks of addiction and potential for abuse. Many Americans are unaware of the addictive nature of opioids, according to a recent public opinion survey which found users overestimate the benefits of prescription opioids and underestimate the risk of addiction and abuse potential (National Safety Council, 2015).

(9)

1.2 Problem Definition & Research Objective

While prescription pain medication has an important place in the medical world when used appropriately for pain relief, the role that prescription opioids have in America's opioid epidemic may suggest that the line between prescription and illicit drugs has become blurred (Smith, 2017). This presents an opportunity to study the opioid epidemic through a sociological lens, by applying a pharmaceuticalization perspective to explore pain medication narratives in a cultural framework. The problem delineation of this dissertation, therefore, is interested in how the pharmaceuticalization of pain in American culture may have normalized prescription pill-taking practices and in turn shaped individual’s attitudes towards pain and risk perception of opioid addiction and abuse potential.

In this regard, this research aims to identify the range of attitudes towards pain and risk perception of prescription opioids among Americans in the U.S. and American Expats in the Netherlands. With this approach, the research is exploratory in nature and poses the question: What are the pain medication narratives among Americans in the U.S. and American Expats the Netherlands? The strategy behind focusing on narratives among these two populations aims to connect individuals’ lives with broader social processes, by contextualizing attitudes towards pain and risk perception of prescription opioids in a cultural framework. A review of relevant literature and existing research on pain and medicine in American culture sets a stage for the theoretical framework of this thesis, which situates pain medication narratives in the context of the opioid epidemic in the U.S. by examining individual’s attitudes towards pain and risk perception of opioid painkillers concerning addiction and abuse potential, by applying concepts of pharmaceuticalization and culture.

(10)

2.0 THEORETICAL FRAMEWORK & LITERATURE REVIEW

2.1 Historical Context of the Opioid Epidemic in the U.S.

For the majority of the 20th century there was a consensus in the medical community that opioids should not be prescribed for chronic pain patients because of the risk they posed for opioid use disorder (OUD), in addition to the lack of evidence concerning the success the drug had for managing this type of pain (Rosenblum et al., 2009). The 1990's was a transformative decade regarding the way in which pain was medicalized in the U.S., with new medical

knowledge, increased pharmaceutical advertising, and changes in the organization of health care as patients began to take on the role of a consumer in regards to having the choice of health insurance policies and medical services (Conrad, 2015).

With the launch of the campaign known as “Pain as the Fifth Vital Sign,” which was backed by the American Medical Association (AMA) and American Pain Society (APS), pain became recognized as a measurable health condition in the same way that blood pressure, heart rate, temperature, and respiration was. This campaign pushed for new guidelines which called on health professionals to do a better job at assessment and treatment of pain and also provided patients with a way to rate their pain experience with the use of a scale ranging from 1 to 10, and specifically advocated for treating pain with prescription opioids (Tompkins et al., 2016). This new conceptualization of pain, which coincided with a “growing recognition in the medical community that many individuals with chronic pain were being treated inadequately” (Bonnie et al., 2017, p. 25; Pokrovnichka, 2008), prompted the American Academy of Pain Medicine and American Pain Society to issue a joint statement outlining the benefits of opioids for chronic pain (Bonnie et al., 2017; (Haddox et al., 1997; Hoffman, 2016). The release of this statement,

(11)

historical context of America's current opioid crisis, which has been argued to have had "unintended consequences resulting in an overreliance on opioids" (Tompkins et al., 2016, p. 13).

Around the same time, pharmaceutical giant Purdue Pharma introduced OxyContin on the market in 1996 as the first extended-release opioid for non-cancer, chronic pain relief, with aggressive yet misleading marketing efforts to doctors and patients alike regarding the drugs' addictive nature and abuse potential. However, the reality was that OxyContin had not been proven to be any safer than the drug Oxycodone, which had already been on the market as a short-acting painkiller with known addictive properties. Among Purdue Pharma’s marketing efforts to encourage the prescribing of OxyContin, the company distributed the discreditable video campaign “I Got My Life Back” to 15,000 physicians across the U.S., promoting the success stories of patients who were using OxyContin for chronic, non-cancer related pain (Lurie, 2017). To note, Purdue did not submit the video to the Federal Drug Administration (FDA) before distributing it to doctors, which is considered a federal law violation. An important feature to note about the misleading nature of OxyContin's marketing is that the company didn't simply omit information regarding the drugs' addiction and abuse potential, but instead addressed the topic head-on. In doing so, they proactively spread the message insisting that opioids are "the best, strongest pain medicine available on the market" and "should be used much more than they are for patients in pain," followed by the claim from Purdue Pharma that "the rate of addiction amongst pain patients that are treated by doctors is much less than 1%" (Goldman & Strickler, 2018). This claim was based on the findings of two prior research studies in 1986 which have since been recognized as misrepresentative and based on insufficient

(12)

2016). These developments have been argued to be influential in shaping attitudes and risk perception among the medical community at this time, resulting in a drastic shift toward more liberalized prescribing of opioids for chronic pain.

Prior to the introduction of OxyContin in 1996, there was reluctance among the medical community to prescribe opioids for anything other than cancer treatment or end-of-life pain management, as U.S. doctors were aware of the highly addictive nature of opioids and tried to avoid treating patients with them whenever possible (Meldrum, 2003). Congress declared 2001– 2011 the “Decade of Pain Control and Research” (Brennan, 2015) which pushed for an increase of recognition of pain as a public health problem (IOM, 2011), in addition to The Federation of State Medical Boards releasing a statement asserting doctors would not receive scrutiny for their opioid prescribing practices, "a fear that had previously reduced the willingness of physicians to prescribe opioids for chronic pain" (Tompkins et al., 2016, p. 13). These have been argued to have produced and reinforced a widespread misconception of the addictive nature of prescription opioids, and they became increasingly used for acute pain i.e. post-surgical and injury-related. Additionally, The Drug Enforcement Agency (DEA) adopted a more "balanced policy" that took a more liberal stance on opioid prescribing in terms of "reducing oversight of physicians that had high rates of opioid prescribing" (Tompkins et al., 2016, p. 13).

It wasn’t until 2007, a decade following OxyContin’s debut on the market, that the U.S. government filed a $635 million lawsuit against Purdue Pharma who pleaded guilty to

misbranding the addictive nature and abuse potential of OxyContin. This was one of the most substantial charges ever to be brought against a pharmaceutical company in U.S. history (Van Zee, 2009). That same year, the growing concern that legal drugs were being used illegally reached its peak when federal government authorities announced that pharmaceuticals had

(13)

become "the most abused illegal drug among young people in the US, behind only cannabis" (Quintero, 2012, p. 6). In 2013, the American Psychiatric Association's Diagnostic and

Statistical Manual of Mental Disorders replaced the categories of "abuse" and "dependence" with the single term "substance use disorder." Building on these efforts, the American Society of Addiction Medicine (ASAM) presented a series of non-stigmatizing language and definitions to replace the former terminology used when discussing prescription drug addiction, abuse and other related conditions (Bonnie et al., 2007). The terminology and definitions presented by ASAM (2011) are as follows:

“Addiction refers to a primary, chronic disease of brain reward, motivation, memory and related circuitry characterized by an inability to consistently abstain,

impairment in behavioral control, craving, diminished recognition of significant problems with one's behaviors and interpersonal relationships, and a dysfunctional emotional response... Dependence refers to a state associated with withdrawal symptoms upon cessation of repeated exposure to a drug. It is important to note that a person who is physically dependent on a drug may not meet the definition of addiction….Tolerance refers to the diminishing effect of a drug resulting from the repeated administration of a given dose...Abuse is no longer acceptable terminology, as research has found the term to be associated with negative and stigmatizing perceptions...The term 'misuse' is commonly used to describe any use of a prescription medication beyond what is directed in a

prescription. It encompasses such specific behaviors and motivations as (1) medically motivated use more frequently or in a higher dose than prescribed, (2) nonmedically motivated use by the person to whom the drug has been prescribed, (3) medical use by a person other than the person to whom the drug has been prescribed, and (4) nonmedical

(14)

use by a person other than the person to whom the drug has been prescribed… Once a patient misusing prescription medication meets the criteria for an opioid use disorder, the term "misuse" is no longer appropriate… Diversion refers to the transfer of regulated prescription drugs from legal to illegal markets. The term is not used in this report to refer to the sharing of drugs with friends, family members, or other contacts for medical or nonmedical purposes. (Bonnie et al., 2017, p. 22-23, Box 1-2; ASAM, 2011)

A majority of the literature in the context of the opioid epidemic in the U.S. has pointed out that rising opioid-related deaths have paralleled an increase in opioid prescribing rates for pain treatment over the past couple decades (Bonnie et al., 2017). The longstanding “war on drugs” in American society, which has focused on policy dealing with illicit drugs, has coincided with aggressive marketing from the pharmaceutical industry to expand markets and sales of prescription medicine through direct-to-consumer (DTC) advertising (Quintero, 2012). It’s been argued that DTC has influenced the doctor-patient relationship, as individuals become more proactive consumers of medical services through increased demands and expectations (Conrad, 2015; Ventola, 2011; McKinley, 2014).

2015 was a major turning point for the opioid epidemic, with new guidelines

administered by the CDC to combat opioid drug abuse alongside federal, state, local efforts to address the prescription drug abuse and heroin use. For treatment of chronic pain (which is defined as lasting longer than three months) the 2016 CDC guidelines recommend using non-opioid treatment whenever possible paired with regularly monitoring patients for harms that outweigh benefits when opioid therapy is indicated (Quest Diagnostics, 2016). Not only is there a lack of evidence that supports the claim that long-term use of opioids is beneficial for chronic

(15)

pain management, but there is data that points to negative patient outcomes such as a heightened risk of OUD and overdose associated with long-term use of opioids. As stated in the 2016 CDC guidelines “long-term opioid use has uncertain [pain management] benefits but known, serious risks” (CDC, 2016b). While these guidelines primarily focus on opioid use and chronic pain, instances of acute pain (i.e. post-surgery or injury related) is equally as important to consider given the fact that evidence has suggested an association between the two, where opioid use for acute pain often leads to long-term use (Dowell et al., 2016). While stricter regulation is a widely proposed solution for reducing the availability of prescription opioids, there is concern that it can have unintended consequences, resulting in the shift from the abuse of one substance to another.

In the beginning of 2016 the U.S. Food and Drug Administration (FDA) released a

comprehensive Opioids Action Plan which paid specific attention to prescription opioids while taking into account the interrelation of prescription opioids use and illicit opioid use, as research suggests the diversion from opioid painkillers to heroin (Bonnie et al., 2017, p. 2). As stated, “prescription and illicit opioid use is intertwined...a majority of heroin users report that their opioid misuse or OUD began with prescription opioids. In addition, the declining price of heroin, together with regulatory efforts designed to reduce harms associated with the use of prescription opioids (including the development of abuse-deterrent formulations), may be contributing to increased heroin use” (Bonnie et al., 2017, p. 6).

2.2 Cultural Theories: Pain & Medicine

The study of the cultural meanings of pain and analysis has been shaped by the influential work of theorists in the field of health and medicine, most namely, Helman (1990), who claimed that "pain responses are shaped and occur within specific social contexts, as opposed to being

(16)

instinctive reactions,” as they "contain a voluntary' component in that action to relieve pain may or may not be sought" (Bendelow & Williams, 1995, p. 157). Following this notion, it’s been said that "responses to pain are predictable on the basis of group membership and that the social meanings ascribed to pain are shared by members of groups" (Freidson, 1970, p. 279-80).

Furthermore, the following has been proposed: "Not all social or cultural groups respond to pain in the same way…How people perceive and respond to pain, both in themselves and others, can be largely influenced by their cultural background…How, and whether, people communicate their pain to health professionals and others, can be influenced by cultural factors" (Helman, 1990, p. 158; Bendelow & Williams, 1995).

American physician Henry Beecher (1959) was one of the first to stress the cultural influence on pain perception and response, arguing that soldiers in World War II who reported minimal or no pain at all associated with their injuries exemplified having an altered pain perception. This, he argued, was due to the context in which they were experiencing pain, as a result of several contextual specific factors such as emotional trauma and being motivated to return home. Cultural theorists Bendelow and Williams (1995) critique the dominant medico-scientific approach to the study of pain for limiting the conceptualization of pain, as it has "resulted in the inevitable Cartesian split between body and mind" (p. 140). In other words, the way in which pain has been medicalized, they argue, has burdened medicine with the task of providing patients with a pain-free experience, which is unrealistic for medicine to be expected to accomplish as it reaches far beyond the realms of its capabilities (p. 160). They go on to discuss the implications of culture on the pain experience, and the importance of "studying narratives as well as the cultural shaping and patterning of beliefs and responses to pain….as it is more than just a medical problem, it's an everyday experience that is shaped by the individual and their

(17)

sociocultural context (p. 140). It’s been argued "the formation of pain beliefs are thought to be a major component in the perception of pain...these beliefs develop through the assimilation of new information with pre-existing meaning and action patterns held by the patient" (Bendelow & Williams, 1995, p. 153; Williams & Thorn, 1989). Furthermore, the following has been

proposed: "1) Not all social or cultural groups respond to pain in the same way 2) How people perceive and respond to pain, both in themselves and others, can be largely influenced by their cultural background 3) How, and whether, people communicate their pain to health professionals and others, can be influenced by cultural factors" (Helman, 1990, p. 158; Bendelow & Williams, 1995).

2.3 Pharmaceuticalization Perspective: Prescription Opioids

Building on a cultural framework for studying pain and medicine in the context of the U.S. opioid epidemic, it’s been argued that pharmaceutical companies may have influenced the medicalization of pain in the U.S. by targeting doctors with promotional marketing as well as public consumers through direct-to-consumer advertising (Conrad, 2015; Williams, Martin & Gabe, 2011). While the concept of medicalization has long been used to study health related topics in the field of sociology and has a variety of definitions, for the interest of this thesis it will be recognized as “the process by which some aspects of human life come to be considered as medical problems, whereas before they were not considered pathological” (Maturo, 2012, p. 2). This notion aligns with the cultural shift reflected in the way in which pain began to take on a new meaning during the twentieth century, where once considered to be a consequential

experience of aging became recognized as a measurable health condition, as previously discussed in regard to the “Pain as the Fifth Vital Sign” campaign, that required medical

(18)

intervention through the use of prescription medication, specifically opioids. (Tompkins, Hobelmann, Compton & Bonica, 2016). It has been observed that there are some aspects of medicalization that do not account for instances in which pharmaceuticals are involved, as stated, “the socio-cultural aspects of pharmaceutical consumption have peculiar features which cannot be properly analyzed by the medicalization framework” (Maturo, 2012, p. 5) and therefore a medicalization framework alone is not always sufficient (Abraham, 2010; Conrad, 2009). For instance, in regard to the interest of this thesis which is centered around the role of

pharmaceuticals in the context of America’s opioid epidemic. Hence, the concept of

pharmaceuticalization has been proposed as a theoretical concept, defined as “the process by which bodily conditions are treated, or deemed to be in need of treatment or intervention, with pharmaceuticals” (Abraham, 2010, p. 290), and will therefore be applied for the purpose of this thesis. Thus, it’s been said that American society has undergone a "pharmaceuticalization of domestic life, where the bedroom and the kitchen are now foci for pharmaceutical marketing and consumption" (Fox and Ward, 2009, p. 857), as stated by The House of Commons (2005) "while the pharmaceutical industry cannot be blamed for creating an unhealthy reliance on, and over-use of, medicines, it has certainly exacerbated it" (Williams et al., 2011, p. 711). In this respect, it can be argued that not only are processes of pharmaceuticalization a means of economic gain by creating a market for pharmaceutical companies to advertise and sell drugs to consumers, but an effort to “reshape consumers’ understanding of their problems into conditions that should be treated by medications” (Horwitz, 2010, p.110; Maturo, 2012). Following this notion, Dr. Art Van Zee (2009), a prescribing physician who specializes in addictive medicine, argues that the marketing of OxyContin was a consistent, systematic effort to decrease risk perception of addiction among doctors and patients alike, by reinforcing the message that “addiction was very

(19)

rare if opioids were used legitimately in the management of pain” (p. 224). Furthermore, DTC has been argued to have promoted a shift in the doctor-patient relationship (Ventola, 2011) as individuals became more “active participants” in their health care (Conrad, 2015; Ventola, 2011; McKinley, 2014). Existing research on physician prescribing practices has suggested that some doctors have a tendency to believe that patients who visit them for pain-related issues deserve to leave with a prescription for pain medication (CASA, 2005). A recent study looked at the influence of patient expectations on physician prescribing practices of opioid painkillers and found that 67% of doctors report that patient expectations influence their prescribing decisions (National Safety Council, 2016). Beyond medical purposes, a pharmaceuticalization lens can also be applied to the recreational use of prescription opioids (i.e. using in order to get high), following the argument that “the problematization of recreational pharmaceutical use may represent a significant departure from these previous practices because it is taking place within a new context – the pharmaceuticalization of society" (Quintero, 2012, p. 14). Furthermore, “the recreational use of prescription drugs is taking place within the context of these broader cultural developments which suggests that pharmaceuticals are increasingly being integrated into

everyday life by a variety of social groups” (Fox & Ward, 2009).

2.4 Stigma: Opioid Use

It’s interesting to apply the theoretical concept of stigma to the context of the opioid crisis, which has been said to have “blurred the formerly distinct social boundary between use of prescription opioids and use of heroin and other illegally manufactured ones” (Bonnie et al., 2017, p. x), specifically in regard to recreational use of prescription opioids and diversion from legal to illegal opioids. According to Ervin Goffman (1963), stigma is “the dehumanization of

(20)

the individual based on their social identity or participation in a negative or an undesirable social category” (Kulesza et al., 2014, p. 2) and is a concept that has been widely discussed in the study of drug use. Existing research has found that "individual drug users attempt to avoid undesirable self-presentation as abusers and reject stigmatized self- identities…and accepting categorical constructions of recreational pharmaceutical use may be problematic to the extent that these representations require individuals to see themselves in a discreditable, stigmatizing light" (Quintero 2012, p. 12.) In the context of the opioid epidemic however, I would argue that it the lack of stigma surrounding prescription opioids that is the problem, which has been brought on by cultural shifts and processes of pharmaceuticalization that have taken place throughout the past few decades in the U.S. and most prominently in the 1990’s. As previously discussed, a new era of pain medicalization and growing acceptance of prescription opioids fueled by the widespread misconception of the drugs addictive nature, which led to the sheer prevalence of prescription pain medication in the everyday lives of Americans — which has been argued to have resulted in increased instances addiction and abuse (from both medical and recreational use). In this regard, recreational prescription opioid users may be able to avoid stigma more easily than illicit opioid users, since they are legal and can be legitimately obtained by a doctor -- unlike illicit opioids such as heroin. For instance, as demonstrated by the series of

“non-stigmatizing” language for prescription drug addiction, abuse and related conditions proposed by the APA and ASAM in 2013, as previously discussed. Moreover, a recent public opinion poll on health attitudes among Americans found that not only did individuals perceive the harms

associated with prescription opioid addiction as being less serious than heroin addiction, despite the drugs chemical similarities, but they viewed heroin users in a more negative, discriminatory light as compared to prescription opioid users (Firth, Kirzinger & Brodie, 2016). This becomes

(21)

problematic due to the fact that despite their legal status, prescription opioids pose the same risk of addiction and abuse potential as illicit opioids. Several researchers have discussed the

digression from prescription to illicit opioids, pointing out that “unlike previous heroin use in the U.S., the nexus of spread is coming primarily out of the medical system, as the line between legal and illicit narcotics has become blurred” (Smith, 2016, p. 1). Furthermore, researchers have discussed the rise of prescription opioid abuse by drawing comparisons to illicit drug use, as they are easier to obtain, more socially acceptable to use, pose less risk of arrest, and are perceived as safer overall (Cicero, Inciardi and Muñoz, 2005). A research study explored recreational drug use among college students and found that using prescription drugs

recreationally was viewed as being a safer, more socially acceptable alternative to using illicit drugs (Bardhi, Sifaneck, Johnson & Dunlap, 2007). This may be due in part to the fact that consumers are more familiar with prescription drugs as opposed to illicit drugs, both directly and indirectly through their own medical use or their observations of family and peers (Quintero, 2009; Quintero & Bundy, 2011; Quintero, 2012), as stated, “some recreational pharmaceutical users categorize prescription drugs they were familiar with as “soft” drugs and see them as relatively safe to consume on physical and social levels” (Quintero, 2012, p. 10). This notion is particularly interesting in regard to the role of prescription medication in America’s opioid crisis, as it may suggest that the socio-cultural context in which individuals come into contact or

interact with prescription opioids as opposed to illicit opioids produces vastly different experiences, and therefore shapes attitudes and risk perceptions.

(22)

In attitudinal research, the concept of persuasion has been said to be “the formation or change of attitudes through information processing in response to a message about the attitude object” (Bohner & Dickel, 2011, p. 403), which is interesting to consider building on the concept of pharmaceuticalization in regards to the role of DTC advertising in America’s opioid epidemic. Bohner & Dickel (2011) discuss the ambivalence nature of attitudes, considering the instance where an individual might have contradictory feelings towards the same object, using cigarette smoking as an example, where "a formerly heavy smoker may have acquired a strong association between the representation of the act of smoking and a positive evaluation; then, on the basis of health information, this person may have formed a new, negative evaluation of smoking, which becomes tagged as valid, whereas the old evaluation persists in memory but becomes tagged as invalid" (Bohner & Dickel, 2011, p. 394). This concept can be applied to the context of the opioid epidemic, for instance, where individuals who rely on prescription pain medication but believe that opioids are highly addictive may experience cognitive dissonance as their beliefs aren’t consistent with their behavior. Additionally, individuals who feel strongly that pain relief doesn’t require prescription opioids, but rely on them when they find themselves in pain and are offered them by their doctors (Clarke, 1999). It’s been argued that attitudes should be regarded as a "fundamental factor" when it comes to studying the way in which people understand and perceive risk (Oltedal, Moen, Klempe & Rundmo, 2004, p. 19). For the purpose of this research, which is interested in pain medication narratives, risk perception of opioid painkillers is mainly of interest since it has been argued in the medical community that the belief that prescription drugs are safe, even when used appropriately, is what leads to instances of addiction and abuse (CASA, 2005). While some theorists argue that attitudes are “stable entities stored in memory” others conceptualize attitudes as being “temporary judgments constructed on the spot from the

(23)

information at hand” (Gawronski, 2007; Bohner & Dickel, 2001, p. 392). This is particularly interesting to think about in regards to the nature of this research when considering the degree to which pain medication narratives are shaped by cultural factors and processes of

pharmaceuticalization. The theoretical framework for my research combined aspects of the range of attitude conceptualizations from both ends of the spectrum, i.e. “stable entities stored in memory” as compared to “temporary judgments constructed on the spot from information at hand” (Gawronski, 2007; Bohner & Dickel, 2001, p. 392), without assuming the sole validity of one or the other. In other words, acknowledging that individuals have stable attitudes to some degree and investigating the extent to which a different cultural context influences their attitudes, if at all.

Two types of risk have been widely discussed among theorists, namely, “objective risk,” which is based on statistical probability i.e. getting struck by lightning, and “perceived risk” which is based on subjective understandings and experiences (Oltedal et al., 2014, p. 11). Risk theorists have proposed several factors that can influence an individual’s risk perception, including "familiarity with the source of danger (Ittelson, 1978), control over the situation (Rachman, 1990), and the dramatic character of the events – rare, striking events tend to be overestimated, while frequency of common events tend to be underestimated (Lichtenstein, Slovic, Fishcoff, Layman & Combs, 1978)" (Oteldal et al., 2014, p. 11). Applying these notions of risk perception to the study of pain medication narratives is interesting in the context of the opioid epidemic. For instance: “familiarity” in regards to the sheer prevalence of prescription pain medication in American society, where and the misconception that they are safe because they are prescribed by a doctor. Secondly, “control over the situation” in terms of individuals are

(24)

expected to be in charge of their own pain management, as seen in the way in which medication is prescribed. For example, it’s been argued that the way in which some doctors administer prescription opioids, i.e., on a “take as needed” basis, may increase risk of addiction and abuse potential (Bonnie, et al., 2017). Lastly, “frequency” concerning the number of incidences of OUD, which could “be argued to desensitized individuals of the risks associated with

prescription opioids. Familiarity is of particular interest in the context of this research, especially when considering that the concept of trust has gained more attention in relation to the study of risk (Oltedal et al., 2004, p. 16). According to Sjöberg (2001) “trust is often held to be of crucial importance for the understanding of risk perception” … “trust in an expert, an agency, or a corporation has been assumed to be determined by perceptions of a number of attributes among them competence and expertise (Peters, Covello & McCallum, 1997; Oltedal et al., 2004, p. 16). In the context of the opioid crisis, trust in doctors, the medical community and the healthcare system at large can be argued to play a role in risk perception of prescription opioids. Combs et al., (2003) identified a set of nine factors that influence subjective risk judgement: “(1)

Voluntariness of risk, (2) immediacy of effect, (3) knowledge about the risk by the person who are exposed to the potentially-hazardous risk source, (4) knowledge about the risk in science, (5) control over the risk, (6) newness, i.e. are the risks new and novel or old and familiar ones, (7) chronic/ catastrophic, that is a risk that may kill people one at a time (chronic risk) or a risk that can kill a large number of people at once (catastrophic), (8) common/ dread, i.e. whether people have learned to live with and can think about the risk reasonably and calmly, or is it a risk that people have great dread for, on the level of a gut reaction, and (9) severity of consequences” (Oltedal et al., 2014, p. 14).

(25)

3.0 METHODOLOGY

3.1 Research Approach

This research used mixed-methods approach to studying pain medication narratives, through analyzing attitudes towards pain and risk perception of prescription opioids. Thus, conducting pilot interviews was used as a way to gather preliminary qualitative insight and inform the construction of the subsequent survey. Additionally, a survey instrument was used to collect quantitative data from likert-item statements in addition to examining qualitative

information through open-ended experiential questions. Research participants were initially recruited through personal and professional networks in the U.S. (Population 1) and online organizations through the Facebook group "American Expats in the Netherlands." (Population 2). Snowball sampling was used to further expand and diversify the two population groups, which totaled to 58 participants (i.e. 38 Americans living in The U.S. and 20 American Expats living in The Netherlands).

As discussed in the earlier chapters, the strategy behind including these two population samples was to account for the cultural context in which pain medication narratives are

embedded. Existing literature has proposed the idea that new experiences may cause an

individual to change or reject existing attitudes (Cross, 2004; Clarke, 1999), as stated, "various researchers have proposed that attitudes are best conceived of as context-dependent, temporary constructions" (Bohner, 2001). For instance, if American Expats who moved from the U.S. and now live in the Netherlands have had new, culturally-specific experiences concerning pain and prescription opioids, they would presumably be required to reflect on their existing prescription pain medicine narratives, which may then differ from Americans who still live in the U.S. who

(26)

ethnomethodological research technique known as “breaching experiments” (Brinkmann, 2016). This method has been used in intercultural research in the social sciences to reveal intricacies of certain practices that are otherwise taken for granted, as stated, “interviewing in different cultural settings as a social practice with a history provides a specific context for human interaction and knowledge production” (Brinkmann, 2016, p. 522). Originally introduced by Harold Garfinkel (1967) in the context of qualitative interviewing, breaching experiments aim to shed light on ”the background expectancies that are rarely thematized in everyday life”(Brinkmann, 2016, p. 523). Including these two populations groups for this research, therefore, aimed to provide a better understanding of the degree to which individual’s pain attitudes and opioid painkiller risk perceptions are culturally-specific and may be reflected in pain medication narratives.

3.2 Pilot Interviews

American Expats living in the Netherlands (Population 2) were connected with through a private Facebook group created for Americans living in the Netherlands, consisting of 622 members at the time research was conducted. In doing so, I posted a message to the group participants where I introduced myself, gave a brief overview of my thesis topic and interest of my research, and asked if anyone would be willing to speak on the phone, Skype or meet in person to talk about their experiences with pain and prescription opioids while living in the Netherlands. I asked willing participants to "like" and/or comment on my post as an indicator that I could contact them to coordinate an informal interview. While this initial attempt to engage with potential participants wasn’t successful in gaining any engagement on the post, I did

however received a few direct messages from group members, all of whom offered to share their experiences over Facebook Messenger instead of scheduling an interview. While I had initially

(27)

planned on conducting these pilot interviews over the phone, on Skype or in person, the lack of responses from my initial attempt and seemingly preferred method of Facebook Messenger led me to re-evaluate my approach for recruiting participants from this population sample. In doing so, I posted a follow-up message in the group offering this method, which resulted in direct messages from select group members who sent me brief accounts of their experiences through directly through the Facebook Messenger platform.

According to social research theories including Brinkmann and Kale (2015), there are three distinct types of interviewers in qualitative research, referred to as the "pollster," the "prober" and the "participant" (p. 528). In regards to the initial qualitative element of my

research (i.e. pilot interviews with American Expats in the Netherlands), I played the role of the "pollster" interviewer type, where I was "primarily seeking the opinions and attitudes of the interviewees" (Brinkmann, 2016, p. 628). In doing so, I presented individuals with an all-encompassing question concerning prescription pain medication experiences, which aimed to capture the essence of attitudes and risk perception through experiential-based accounts before designing and conducting the survey model that was later used. To note, while these pilot interviews were initially intended to solely serve as an informative, messaging resource for the construction of the Likert-item statements for the survey, they ended up being re-examined and used as an additional means of qualitative research due to the shortcomings of the quantitative analysis from the survey. Key findings from these interviews are presented and discussed in the following chapter.

(28)

Quantitative data was collected through the online survey platform, Typeform (https://www.typeform.com/), which was used to explore a variety of attitudes towards statements concerning pain and prescription opioids. In doing so, the majority of the survey consisted of likert-item statements, a research technique developed in 1932 by Rensis Likert to measure attitudes, typically consisting of a 5- or 7-point ordinal scale which asks respondents to indicate the degree to which they agree or disagree with certain statements (Sullivan, 2013). While responses in an ordinal scale can be ranked, it's not possible to measure the exact distance between them i.e. the distance between ‘strongly agree' and ‘agree' might not mean the same thing as the distance between ‘agree' and ‘neutral' (Sullivan, 2013). By using a Likert item type format, I aimed to identify patterns of acceptance and disagreement accounts across two

population samples, which may reveal similarities and differences between and within the two cultural contexts. The final survey consisted of 26 total items, including three demographic questions (i.e. location, age and gender), 21 Likert-item statements concerning pain, opioids and prescription pain medication, in addition to two experiential questions with open-ended

responses regarding participant’s firsthand and/or second-hand experience with prescription opioids and addiction/abuse.

The Likert-item statements asked research participants to indicate the extent to which they agreed/disagreed with each statement using a 5-point Likert scale (1=strongly disagree; 2=disagree; 3=neutral; 4=agree; 5=strongly agree). To control for acquiescence bias as much as possible, which is when respondents have the tendency to provide affirmative answers to

questions (Billiet et al., 1998; Hinz et al., 2007), the statements were framed in a way that balanced both positive and negative wording towards the attitude objects i.e. pain, opioids, and prescription pain medication. In doing so, the full set of 21 likert-item statements composed of

(29)

three distinct subgroups, including: 1) pain-specific 2) opioid-specific and 3) prescription pain medication, generally. This is discussed and illustrated in more detail in the following chapter. In addition to avoidance of acquiescence bias regarding the way in which participants responded to the survey statements, it’s been argued that when statements in survey research are all framed either positively or negatively -- lacking a balance between the two -- participants are less compelled to pay close attention when reading the statements which may result in

misrepresented data (Swamy, 2007, p. 496-505).

The process of constructing the statements involved the assessment of various sources regarding pain, opioids and prescription pain medication, including relevant literature and theoretical concepts, existing research studies, public forums and media representations, as well as the experiential-based messaging gathered from the pilot interviews. [SEE VISUAL 1].

VISUAL 1. Likert-item statements

# SURVEY STATEMENTS

1 Pain is part of the healing process

2 I’d rather be prescribed pain medication and end up not needing/using it then not receive it at all

3 I worry about the risk of addition with opioid painkillers, even when they are using as directed by a doctor

4 Pain is an inconvenience

(30)

treatment

6 Opioid painkillers are more often a quick fix rather than the most effective treatment

7 In most cases with prescription opioids, the benefit of pain relief outweighs the risk of addiction

8 It’s okay to use someone else’s pain medication, as long as it’s intended for medical purposes

9 After an injury or surgery, I would only use opioid painkillers if over-the-counter pain medications (i.e. ibuprofen) weren’t strong enough

10 If I’m in pain I expect to receive prescription medication from my doctor

11 Anyone can become addicted to opioid painkillers, even when used as directed by a doctor

12 Sharing prescription opioids with a friend or family member is okay if they are in pain

13 Recreational use of prescription painkillers is less dangerous than use of illegal drugs such as cocaine

14 Prescription opioids can easily lead to heroin use

15 I’d rather be in pain after an injury or surgery than worry about becoming addicted to opioids

16 It’s more socially acceptable to use prescription painkillers recreationally than to use most hard drugs

(31)

17 Prescription opioids are essential for managing chronic pain that would otherwise be unbearable

18 It can be useful to keep leftover/unused prescription painkillers in the event of an injury

19 More often than not, pain relief requires prescription medication

20 Prescription opioid addiction is equally as serious as heroin addiction

21 Pain isn’t always a bad thing

As shown above, statements 13, 14, 16 and 20 focus on risk perception of addiction and abuse related to prescription opioid painkillers in relation to use of illegal drugs. These

statements were constructed in a way that reflects the notion of familiarity in regards to how users classify prescription drugs (Quintero, 2012) as well as the previously cited research study findings pertaining to college student’s perceptions of recreational drug use being safer and more socially acceptable that illicit drug use (Bardhi, Sifaneck, Johnson, & Dunlap, 2007). Also, based on the widespread belief that prescription opioid abuse is a significant risk factor for heroin use, (Rudd et al. 2016; Compton et al., 2016) which is supported by data from the CDC (2017) that found the majority of heroin users report prior abuse of prescription opioids, in addition to the argument that "the line between legal and illicit narcotics has become blurred" (Smith, 2017). Statements 8, 12 and 18 reflect the argument that while the majority of

individuals who are prescribed opioids may not end up addicted or develop OUD, prescription opioids can have harmful implications not only to the individuals who are prescribed them but can have harmful societal consequences, as they can "make their way into the hands of people

(32)

debate surrounding the risk/benefit assessment when prescribing opioids for pain relief, citing the lack of evidence concerning the success of using prescription opioids for chronic pain in contrast to the existing evidence that suggests the potential risks associated with long-term use (Rosenblum et al., 2009; Bonnie et al., 2017). Additionally, in regards to recent findings from a public opinion survey that suggested many individuals not only unaware of opioids risk of addiction and abuse potential addictive nature of opioids but tend to overestimate the benefits of prescription opioids (National Safety Council, 2015). Statements 3, 5, 11 and 20 follow the theoretical notion that risk perception can be considered to be culturally constructed and produced through social processes of attribution and framing, for example, through that of prescription drugs as opposed to that of illicit drugs, as stated by (Brown and Calnan, 2012). Similarly, the role of trust in doctors and the medical system in regards to risk perception of prescription opioids (Oltedal et al., 2004, p. 16), as well as the findings from a recent poll that suggest Americans perceive prescription opioid addiction as being less serious than heroin addiction (Firth et al., 2016). Statement 9, 15 and 17 are based on cases in which using

prescription opioids for acute pain (i.e. post-surgery or injury) can become problematic when it leads to long-term opioid use, as stated, “when opioids are used for acute pain, clinicians should prescribe the lowest effective dose of immediate-release opioids and should prescribe no greater quantity than needed for the expected duration of pain severe enough to require opioids...three days or less will often be sufficient; more than seven days will rarely be needed” (CDC, 2016a). Also, in consideration of the claim that opioid pain medication isn't' always more effective than non-opioid medicines for pain, and therefore should not be the first-line treatment (CDC, 2016a).

The two experiential questions at the end of the survey aimed to gain insight beyond the data provided by the Likert-item statements, asking: “Have you ever been prescribed an opioid

(33)

for pain?” and “Have you or anyone you know experienced addiction as a result of being

prescribed opioids for pain relief and/or using recreationally?” For the purpose of this research, I focused on the knowledge-based factors that influence risk judgement, previously cited by Combs et al. (2003) and Oltedal et al. (2014), as demonstrated in the open-ended question in the survey that asked respondents whether they are someone they knew had experienced prescription opioid addiction. This was a way to explore whether this type of knowledge was an indicator of a higher score of risk perception, assuming that beyond exposure to scientific-knowledge of risk of addiction and abuse potential of prescription opioids these individuals had firsthand and/or second-hand knowledge of the risk source. In order to gain deeper insight to this possible

influential factor, respondents who reported "yes" were asked to share details of their experience in an open-ended format.

4.0 DATA ANALYSIS & RESULTS

4.1 Summary

As mentioned in the previous chapter, using a mixed-methods approach aimed to allow for the connecting of the quantitative survey data with the qualitative experiential information that was collected, in efforts to allow for a more holistic understanding of the pain medication narratives that were identified among both population samples. Analysis of both the quantitative and qualitative data from this research is discussed is more detail in the following sections of this chapter.

(34)

method proposed by Billiet et al (1998), as stated "the responses to the negative items are in reverse order coded and then added to the responses of the positive items to give a single score as a measure of the attitude construct…Acquiescence to the negative items will offset acquiescence to the positive items and consequently the mean of the response distribution will not be biased" (p. 130). The reason being, as mentioned previously, were that the statements were framed with both positive and negative language to avoid acquiescence bias (i.e. more favorable towards pain, opioid and prescription pain medication). Statements 2, 4, 5, 7, 8, 10, 12, 13, 17, 18, 19 and 20 (Group A) were framed in a way that had a more positive slant towards using opioids for pain relief with less concern about the risk of addiction. In contrast, statements 1, 3, 6, 9, 11, 14, 15, 16, and 21 (Group B) were framed in a way that was more opposed to using prescription opioids for pain relief with more caution about the risk of addiction and abuse potential. While the balance in the positive and negative framing of the statements was deliberate and important for the survey process, as it allowed for most accurate and unbiased responses, the statistical data that was produced needed to then be transformed in order to allow for accurate interpretation. Using the "Recode Variables" function in SPSS, the Likert-item values for the statements in Group A were reversed in order to be consistent with the statements Group B. In doing so, the value score of 1 (strongly disagree) was re-coded to a value score of 5 (strongly agree) and vice versa; a value score of 2 (disagree) was re-coded to a value score of 4 (agree) and vice versa; and the value score of 3 (neutral) remained the same (IBM Corp, 2017). The reason for this decision was to allow for a way of interpreting the quantitative data in a more qualitative way in order to understand what narratives emerged from the way in which participants responded to the likert-item components, specifically when comparing the means scores of the statements across the two population samples. In other words, a higher score would suggest a more positive attitude

(35)

towards pain and stronger risk perception of opioid addiction and abuse potential. In contrast, a lower score would indicate a less positive attitude towards pain and a weaker risk perception of opioid addiction and abuse potential.

Before conducting any form of analysis, two distinct types of pain medication narratives were expected to emerge from the quantitative survey data -- likely falling on opposite ends of the spectrum in terms of likert-item mean scores reflective of individual’s attitude towards pain and risk perception of opioid addiction and abuse. The first, based on a higher mean scores, would describe a narrative that is more opposed the using prescription opioids for pain relief (Narrative 1) . The second, based on a lower mean score, would indicate a type of narrative that is more accepting of prescription pain medication (Narrative 2). Since these are only two

possible narratives -- and are in complete opposition of one another -- the survey statements were constructed in a way that allowed for a deeper understanding of the underlying nature of

narratives and distinct aspects that comprised of attitudes and risk perception and would likely inform additional pain medication narratives. In doing so, three of the statements investigated attitudes towards pain in general (statements 1, 4 and 21), four of the statements assessed risk perception of prescription and illicit opioids (statements 13, 14, 16 and 20) and the remaining 14 statements explored the range of attitudes and risk perception pertaining to prescription opioids and pain relief. [SEE VISUAL 2].

VISUAL 2. Likert-item subgroup statements

# STATEMENT SUBGROUP

1 Pain is part of the healing process PAIN

(36)

needing/using it then not receive it at all MED

3 I worry about the risk of addition with opioid painkillers, even when they are using as directed by a doctor

PRESCRIP PAIN MED

4 Pain is an inconvenience PAIN

5 If my doctor prescribes an opioid painkiller I trust that it is the most appropriate treatment

PRESCRIP PAIN MED

6 Opioid painkillers are more often a quick fix rather than the most effective treatment

PRESCRIP PAIN MED

7 In most cases with prescription opioids, the benefit of pain relief outweighs the risk of addiction

PRESCRIP PAIN MED

8 It’s okay to use someone else’s pain medication, as long as it’s intended for medical purposes

PRESCRIP PAIN MED

9 After an injury or surgery, I would only use opioid painkillers if over-the-counter pain medications (i.e. ibuprofen) weren’t strong enough

PRESCRIP PAIN MED

10 If I’m in pain I expect to receive prescription medication from my doctor

PRESCRIP PAIN MED

11 Anyone can become addicted to opioid painkillers, even when used as directed by a doctor

PRESCRIP PAIN MED

(37)

12 Sharing prescription opioids with a friend or family member is okay if they are in pain

PRESCRIP PAIN MED

13 Recreational use of prescription painkillers is less dangerous than use of illegal drugs such as cocaine

OPIOIDS

14 Prescription opioids can easily lead to heroin use OPIOIDS

15 I’d rather be in pain after an injury or surgery than worry about becoming addicted to opioids

PRESCRIP PAIN MED

16 It’s more socially acceptable to use prescription painkillers recreationally than to use most hard drugs

OPIOID

17 Prescription opioids are essential for managing chronic pain that would otherwise be unbearable

PRESCRIP PAIN MED

18 It can be useful to keep leftover/unused prescription painkillers in the event of an injury

PRESCRIP PAIN MED

19 More often than not, pain relief requires prescription medication PRESCRIP PAIN MED

20 Prescription opioid addiction is equally as serious as heroin addiction

OPIOIDS

(38)

For quantitative data analysis, descriptive statistics of survey respondents and results were calculated, Regression and T tests were run, and Cluster Analysis was carried out using the Statistical Package for Social Sciences (SPSS) software program (IBM Corp, 2017). Descriptive statistics play an essential role in quantitative analysis as they set the foundation of the study by summarizing the key features of the data sample and measures (Statistical Solutions, 2018b). Cluster analysis was used as a means of identifying smaller groups that share similar

characteristics within a larger group of cases in regard to certain variables (Mooi & Sarstedt, 2011, p. 238), and aimed to group survey respondents across both populations based on similarity patterns in regards to certain variables of interest that had the most significance in predicting cluster membership.

Of all the survey respondents (n=58), the descriptive statistics for the demographic data was recorded as follows: 66% Americans living in the U.S. (n=38) and 34% American Expats living in the Netherlands (n=20); 62% females (n=36) and 38% males (n=22); 16% ages 18-24 years (n=9), 38% ages 25-34 years (n=22), 5% ages 35-44 years (n=3), 14% ages 45-54 years (n=8), 16% ages 55-64 years (n=9) and 12% ages 65-74 years (n=7). Of Americans in the U.S., 61% of respondents were between the ages 18-34 years and 39% were between the ages of 45-74 years. No respondents were in the 35-44 year age range. Of American Expats living in the

Netherlands, 55% of respondents were between the ages 25-44 years, while 45% were between the ages of 45-74 years. No respondents were in the 18-24 year age range. As a result of the missing age range for each of the population samples, it can be concluded that the average age of respondents in Population 1 (Americans in the U.S.) was slightly lower than that of Population 2 (American Expats in the Netherlands). For both populations, the majority of respondents were 25-34 years old. Both populations were similar in terms of the gender of respondents, with

(39)

females accounting for 61% of respondents among Americans in the U.S. and 55% of respondents among American Expats in the Netherlands (IBM Corp., 2017).

When it came time to analyze the data , T Tests and Regression tests were run to compare the means of both population groups to see which variables were most useful for making an educated guess about whether there was a significant correlation between a respondent's location (i.e. the U.S. or the Netherlands) and their attitude towards pain and risk perception of

prescription opioids, while also controlling for age and gender in some instances to further investigate as needed. In other words, these tests aimed to address whether American Expats in the Netherlands had different attitudes towards pain and risk perception of opioid painkillers than Americans in the U.S.

First, an Independent Sample T-Test was run to calculate the probability of association with group membership by determining whether there is a significant, statistical difference between two population means (Statistical Solutions, 2018c). This type of test compared the average of the likert-item mean scores between the two population groups, defined by the location variable i.e. Americans in the U.S (Population 1) and American Expats in the

Netherlands (Population 2) to determine whether the average difference between the two groups was significant or if it was instead due to random chance. In other words, the mean scores of the likert-item data produced by both population groups were calculated and compared to see if there was any indication that American Expats in the Netherlands have different attitudes towards pain and risk perception of opioid painkillers than Americans in the U.S. The dependent variable was the likert-item mean scores (attitude/risk perception) and the independent variable was location (American Expats in the Netherlands/Americans in the U.S.). First, a T-Test was run for the comprehensive mean score (inclusive of the 21 likert-item response data) of the two population

(40)

groups, revealing almost statistically identical scores between the U.S. (4.1) and the Netherlands (4.0) with a T-test outcome of 13% probability (0.13). Due to this lack of statistical difference, analysis then narrowed in on the mean scores for each of the three subgroups mentioned above (i.e. pain-specific, opioid-specific, and prescription pain medication) separately, to see if this indicated any difference based on location. In doing so, the biggest difference in mean scores was found in the responses to the opioid-specific statements, as the U.S. ranked higher (4.6) than the Netherlands (4.0) with a T-test outcome of 6% probability (0.06) which might indicate a stronger risk perception of opioid painkillers. Next was the pain-specific subgroup, where the Netherlands ranked slightly higher (4.1) than the U.S. (3.9) with a T-test outcome of 14% probability (0.14), which could suggest a more positive attitude towards pain. Lastly, for the prescription medication subgroup the U.S. ranked minimally higher (4.1) than the Netherlands (4.0) with a T-test outcome of 24% probability (0.24). While there was some degree of statistical differences in subgroup mean scores between the two population groups, which could possibly suggest a more positive/negative pain attitude and stronger/weaker risk perception of opioid painkillers, this wasn’t enough evidence to assume that it was solely due to location -- random chance or other unknown variables must be considered possible influencers.

With this consideration, a Binary Logistic Regression test was then run to try to get a better understanding of the extent to which the location variable (population group membership) could be predicted based on the likert-item survey data (i.e. responses to statements regarding pain attitudes and risk perception of opioid painkillers), while also controlling for variables including gender and age group. For each subgroup, several statements were statistically noteworthy. When analyzing the pain-specific subgroup while controlling for location and gender, Statement 4 (“Pain is an inconvenience’’) had a strong, negative correlation (-1.402)

(41)

with high significance (.002). When analyzing the opioid-specific subgroup while controlling for the demographic variables (location, gender and age range) Statement 14 (‘’Prescription opioids can easily lead to heroin use’’) had a medium, negative correlation (-.629) with high significance (.037). When analyzing the prescription pain medication subgroup while controlling for location, gender and age range the statements that had the most statistical significance were as follows:

Statement 3 (‘’I worry about risk of addiction with opioid painkillers, even when they are used as directed a doctor’’) had a strong, positive correlation (11.403) with high

significance (.028)...Statement 6 (‘’Opioid painkillers are more often a "quick fix" as opposed to the most effective treatment’’) had a strong negative correlation (-2.239) with high significance (.022)...Statement 17 (‘’Prescription opioids are essential for managing chronic pain that would otherwise be unbearable’’) had a strong negative correlation (-2.048) with high significance (.016)...Statement 18 ‘’(It can be useful to keep

leftover/unused painkillers in the event of an injury’’) had a strong negative correlation (-1.845) with low significance (.067).

As a final step in quantitative analysis, a clustering technique was used in efforts to identify the existence of subgroups within the larger group of all research participants and to what degree cluster membership was based on the variables of interest, most namely location and attitudes/risk perception (i.e. likert-item scores), while also taking into account gender and age. In accordance with the exploratory nature of this research, which is interested in pain medication narratives i.e. whether American Expats in the Netherlands have different attitudes towards pain

Pain medication narratives : exploring attitudes towards pain and risk perception of opioid painkillers among Americans in the U.S. and American expats in the Netherlands

Pain Medication Narratives

Exploring Attitudes Towards Pain and Risk Perception of Opioid

Painkillers Among Americans in the U.S. and American Expats in the

TABLE OF CONTENTS

1.0 INTRODUCTION

1.1 Contextual Relevance & Existing Research

2.0 THEORETICAL FRAMEWORK & LITERATURE REVIEW

2.1 Historical Context of the Opioid Epidemic in the U.S.

2.2 Cultural Theories: Pain & Medicine

2.3 Pharmaceuticalization Perspective: Prescription Opioids

2.4 Stigma: Opioid Use

3.0 METHODOLOGY

3.1 Research Approach

3.2 Pilot Interviews

4.0 DATA ANALYSIS & RESULTS

4.1 Summary