The Controlled Substances Act of the United States regulates the use of drugs by placing them into 1 of 5 descending schedules ranging from schedule I (substances with high potential for abuse or dependence and no accepted medical use in the United States) to schedule V (low potential for abuse or dependence and an accepted medical use in the United States). Tapentadol is a centrally acting, atypical analgesic with a novel mechanism of action that is a combination of μ-opioid agonist activity and norepinephrine reuptake inhibition used to treat moderate to severe acute and chronic pain [1-3]. It was placed in schedule II by the Drug Enforcement Agency because it was determined to have a high potential for abuse, an accepted medical use in treatment in the United States, and the possibility of leading to severe psychological or physical dependence [4]. Tapentadol immediate-release ([IR]; Nucynta) was approved by the United States Food and Drug Administration in December 2008 and the extended-release (ER) formulation (Nucynta ER) in August 2011 [5]. The abuse liability of tapentadol has been of interest since its release, in part because of its proposed mechanism of action and in part because of its initial identification as a schedule II opioid.

Thus far, the prevalence of tapentadol abuse and diversion (both ER and IR) has been characterized as low compared with other prescription opioid compounds, particularly when considered at the population level [6-8]. For instance, from the fourth quarter of 2011 to the second quarter of 2016, tapentadol had an event rate of 0.015 for intentional abuse, an event rate of 0.029 for diversion, and an event rate of 0.245 for past 30-day use to get high. Comparator opioid active pharmaceutical ingredients (hydrocodone, hydromorphone, morphine, oxycodone, oxymorphone, and tramadol) were reported as being intentionally abused from 7.41 (oxymorphone) to 84.32 (oxycodone) times the rate of tapentadol intentional abuse, diverted from 23.172 (oxymorphone) to 316.862 (oxycodone) times the rate of tapentadol diversion and used for getting high in the last 30 days from 3.48 (tramadol) to 52.97 (oxycodone) times the rate of tapentadol [9]. Tapentadol is associated with the fewest serious adverse events among comparator prescription opioid active pharmaceutical ingredients, as well as the fewest dosage units and prescriptions dispensed in the United States [10]. It has also been associated with a lower risk of seeking out multiple physicians to provide prescriptions than oxycodone [11-13]. Nevertheless, when adjusted for prescription volume or drug availability, low levels of abuse-related outcomes are consistently present [9,14].

It is difficult to make specific inferences about tapentadol abuse or nonmedical use (NMU) because little has been published on user experiences. This is complicated by the comparatively low rates of tapentadol dispensing in the United States [10], low rates of internet posting about tapentadol compared with other prescription opioids [15], and few behavioral pharmacological studies [16]. Therefore, this study sought to address this gap by piloting web-based recruitment for a tapentadol NMU survey.

The Tapentadol Use Internet Survey (TUIS) is a web-based survey followed by an interactive chat among interested survey completers (Multimedia Appendix 1). The survey queried the motivation for tapentadol NMU, sources of drug procurement, routes of administration, tampering methods, doses used, and impressions of tapentadol products (Nucynta and Nucynta ER). The goal of the survey was to solicit detailed information from individuals who self-reported lifetime tapentadol NMU and to pilot the use of web-based recruitment for the difficulty of finding research participants.

NMU was defined as the use of tapentadol “in a way not prescribed,” including any of the following: (1) used if not prescribed to you; (2) used for reasons other than as a treatment for pain; (3) used via an alternate route of administration (eg, snorted, injected, or other routes not intended for the product); (4) used after tampering (eg, crushed); (5) used in combination with alcohol, illicit drugs, or other prescription drugs without doctor approval; or (6) used at a higher dose than prescribed.

Bluelight.org [DragonByte Technologies Ltd] was selected from a pool of drug discussion websites to host the TUIS because at the time of this study its culture supported authentic posts and harm reduction, the site reported a range of approximately 7000 to 10,000 active users within any 30-day period [17], posts were in English, the site was considered stable because it had been in existence for over 10 years, and the staff encouraged research collaborations. A growing number of studies have included data collected from Bluelight.org [18-23].

The 36-item TUIS was developed to solicit information describing tapentadol NMU. Survey construction was an iterative process whereby investigators designed questions to elicit information about participants’ prescription opioid use history, and, in particular, their experience with and impressions of tapentadol when used nonmedically, the motivation for tapentadol NMU, sources of procurement, routes of administration, tampering methods, doses used, and general impressions of tapentadol products. The survey was reviewed by Bluelight moderators to ensure their research standards were met [24]. After the investigative and Bluelight teams agreed on content, the survey was posted on Bluelight.org.

At the completion of the TUIS, participants were asked to consider chatting interactively with a researcher about their tapentadol product NMU. The chat consisted of 14 questions that provided a framework to enable probing for individual details about initial tapentadol exposure, first use of tapentadol, the high associated with tapentadol, the formulation of tapentadol that had been used for NMU, why it had been used for NMU, number of times using, and the dose of tapentadol most often used for NMU.

Individuals participating in the TUIS had to be at least 18 years of age, able to read and understand the English language, reside in the United States, and be willing to provide consent to participate in the survey. They also had to have visited Bluelight.org and report lifetime NMU of a tapentadol product.

Individuals participating in the postsurvey interactive chat had to have completed the TUIS. They had to be willing to provide contact information (email address or Bluelight username) for chatting purposes. Participants had to have the ability to use a web-based chat program and to provide consent to participate.

The study was approved for conduct by the New England Institutional Review Board (NEIRB 120170005: Internet Survey and Online Chat Interviews Regarding Tapentadol Use).

Participants were recruited for TUIS completion from January through May 2017. A recruitment flyer and banner advertisement were placed on the Bluelight.org website with a link to the web-based survey. This link directed individuals to the consent page which described the voluntary nature of the survey, the absence of payment for survey completion, and information about how to complete the survey. Selecting I agree to participate on the informed consent page moved the participant to the beginning of the survey, whereas individuals who did not provide consent to participate in the survey were thanked for their time and brought to an end page. Survey completion took between 5 and 20 minutes depending on the detail of responses provided.

Upon completion of the survey, participants were asked if they would like to be considered for a postsurvey interactive chat regarding their use of tapentadol products. Participants were guaranteed anonymity and asked to provide an email address or a Bluelight username, so they could be contacted by a member of the research team to set up the chat. At the conclusion of the approximately 1-hour, semistructured chat, participants were offered the choice of receiving a US $25 Amazon.com gift certificate or donating the same amount to Erowid.org (a nonprofit drug-education website).

TUIS responses were collected using the web-based data collection software Qualtrics (Qualtrics) and stored in a secure database. The survey was designed for sequential completion such that it was necessary to complete particular items of interest before moving forward in the survey (forced choice), and responses to previously answered questions were carried forward when requesting more detail. A Qualtrics technology function that blocked more than one survey per individual was enabled. Survey items were examined descriptively with frequency and percentages for categorical and binary variables, and means, medians, error, and ranges for continuous variables. Data analysis was carried out using SAS (version 7.11; SAS Institute).

The postsurvey interactive chat was conducted using Cryptocat, a free, open source, encrypted web-based chat program. Transcripts from the postsurvey interactive chats were saved on secure servers with access permissions given only to research staff for data analysis. Thematic data analysis based on grounded theory qualitative research methodology was used to analyze the interview data [25-27]. This is an analytic approach whereby the raw data drive thematic development and analysis.

Specifically, reviewers conducted in-depth review of the chat content to discern and characterize emerging themes regarding tapentadol NMU. In this open coding phase, 2 reviewers (TDG and JB) read the first half of transcripts and assigned topic categories or themes. Constant comparative analysis was performed by repeatedly going back and reviewing previously established categories. While reviewing transcripts, reviewers continually evaluated whether a certain thought or response from an interview participant fit into a previously existing category or whether it represented a unique new theme. This round of review established a basic coding structure.

Subsequently, the 2 reviewers (TDG and JB) read the remaining transcripts and assigned topic categories or themes as appropriate. If additional categories were discerned, the previous transcripts were revisited to ensure that all categories were adequately captured. In this axial category round of analysis (axial coding uses the predefined concepts and categories while rereading the text to confirm that they accurately represent interview responses and to determine how they are related), reviewers established how the categories related to one another. Finally, final selective coding established overall hypotheses and explanations regarding reasons for, opinions of, and experiences with tapentadol NMU.

A total of 8 survey participants (10% of the survey sample) agreed to participate in a follow-up semistructured interactive web-based chat. The demographic profile was similar to that of the full sample: 7 (88%) were male, 6 (75%) were White, 6 (88%) were aged <27 years, and 7 (88%) attended a minimum of some college. All participants reported tapentadol IR NMU and half (4/8, 50%) reported tapentadol ER NMU. A total of 6 (75%) participants reported prescription opioids or heroin as their preferred drug for NMU. The remaining 2 (25%) participants reported marijuana and psychedelics (n=1) and dissociative drugs (n=1) as drugs of choice.

Table 9 summarizes the themes and supporting statements derived from participants’ descriptive responses to questions regarding their tapentadol NMU experience. Sentiments and side effects were coded positive if they were favorable or neutral if participants indicated that they neither liked nor disliked the experience or liked one aspect but disliked another. Negative experiences were coded as such if the experience was clearly disliked and typically included extreme or no effects at higher doses.

Of the 8 participants, 3 (38%) reported favorable or positive sentiment, 3 (38%) reported neutral sentiment, and 2 (25%) reported negative sentiment. The side effects of tapentadol were described solely in neutral to negative terms. Reported tampering efforts reflected the composition of the ER formulation (Nucynta ER is formulated with inactive ingredients that make it difficult to crush, although it is not recognized by the Food and Drug Administration as having abuse-deterrent properties): “Nucynta [ER] was like a solid chunk of plastic.” They also reflected the rationale for the oral use of both ER (“I could barely change its shape in the slightest [and] ended up swallowing it whole”) and IR formulations (“If I don’t need to beat a time-release formulation, I typically just swallow 95% of the time”). Comments also reflected the undesirability of the IR formulation for alternate routes of administration: “We actually all tried to snort it… [but] it burned and hurt superbly” (Table 8). Finally, tapentadol was referred to as a fairly obscure find in traditional diversion settings. An individual reported that when trying to sell it or even share it, people “wouldn’t have anything to do with it” because they did not recognize the product.

To date, studies on the abuse liability of tapentadol have focused on aggregate outcome measures [6,7,9,14,15,28]. The present survey sought to address a gap in the tapentadol literature by soliciting direct feedback from individuals with tapentadol NMU experience and characterize the associated motivations, behaviors, and consequences of NMU. To do so, web-based recruitment and survey technology were piloted, and it was found that they were an effective method to recruit a difficult-to-find research sample.

Similar to other prescription opioids [29], the main source of tapentadol was friends, family, or acquaintances. Tapentadol was also obtained on the web and through other sources of diversion, such as being stolen, drug dealers, or purchased directly from friends or family. Although many sources have not reported significant levels of diversion [7,9,14,15], these data reveal a type of tapentadol diversion that is occurring, although perhaps at low levels.

It was hypothesized that individuals might use tapentadol for reasons other than analgesia, such as the rumored psychedelic effects [5,30,31] (Tables 6-8). However, the primary reason for tapentadol NMU and ongoing tapentadol NMU across both formulations was better pain relief, followed by psychotropic effects, including relaxation, reduction in depression or anxiety, or getting high. Approximately 25% to 30% of participants reported that they misused their own tapentadol prescription, revealing NMU among some patients with pain. Pain has been found to contribute to the risk of developing prescription opioid use disorders over time [32], and it is possible that the present data capture aspects of this relationship.

Benzodiazepines are most often used in combination with both formulations of tapentadol. Benzodiazepines can pose a life-threatening risk when used concomitantly with opioids because of the increased risk of respiratory depression and overdose [33,34]. Even so, they are prescribed at varying rates to patients undergoing opioid maintenance therapy. Individuals seeking opiate detoxification also report using benzodiazepines to manage anxiety, help with sleep, decrease opioid withdrawal, enhance the recreational effects of other drugs or substances, or get high [35]. The rationale for the concomitant use of benzodiazepine and tapentadol was not discussed in this study.

Individuals who used tapentadol at high doses (≥200 mg) reported hallucinations. Some interview participants did not identify these effects as positive or negative, but others reported them as strong deterrents to future tapentadol NMU. Hallucinations are described as part of a serotonin syndrome that can occur when taking serotonin and norepinephrine reuptake inhibitor products such as tapentadol [5] and in combination with other serotonergic drugs [36]. To date, the literature is inconclusive as to whether tapentadol alone has resulted in a true serotonin syndrome experience [30,31].

Desirability ratings were lower for tapentadol than for the other opioid compounds. Ratings of desirability were similar between tapentadol ER and IR, with tapentadol ER being less attractive for NMU than IR. In addition to supporting recent findings that IR formulations are more desirable for NMU than ER formulations [37], these data also suggest a lack of desirability for the entire tapentadol molecule, not just for one formulation. In further support of this inference, most participants (61/78; 78%) stopped using either formulation at the time of the survey. These findings were similar to those reported in the study by McNaughton et al [15]. Regardless of how individuals reported using tapentadol, most participants (ER 22/30, 73%; IR 55/67, 82%) did not indicate that they enjoyed it.

Limitations of this study include the use of self-reported data from a self-selected, US-based, convenience sample responding to a pilot internet survey, which may not be a fully representative sample of nonmedical users of tapentadol. The selection of individuals who reported lifetime tapentadol use meant that some did not report their current experiences. The modest number of participants may be because of the market share of tapentadol, but it may also be because of the nondesirability of tapentadol for NMU. This may be a topic for future research [8]. Extension of the recruitment period longer than 5 months might result in a larger sample size, if the reason for the modest sample is the lack of tapentadol market penetration. Notably, few participants volunteered to participate in the follow-up survey. Although it is possible that this activity may not have been of interest, it may also have been because of the requirement to provide an email address or a Bluelight username (which suggests indirect support for anonymous web-based surveys). However, it also suggests an opportunity for another technological development in which survey completers could remain anonymous yet respond to specific follow-up questions. Much of what was reported is similar to other findings in the literature documented herein, lending face value to this report. New directions for future surveys may include patterns of tapentadol NMU, such as frequency of use, redosing, and the degree to which larger doses are used to obtain the same effect. Finally, participants were also aware that the focus of the study was tapentadol and may have felt compelled to over- or underreport its use. Even so, great care was taken to ensure the quality of data collection and analysis.

In conclusion, these preliminary data reveal potential avenues for further exploration of NMU tapentadol. The use of web-based survey technology for survey recruitment of a difficult-to-find sample and a follow-up interactive chat may be another useful technology for postmarketing surveillance studies. The primary motive for continuing tapentadol NMU was pain relief. Tapentadol ER (12/21, 57%) or IR (28/47, 60%) use together with benzodiazepines were reported . There is also some evidence of diversion. At high doses, psychotropic effects have been reported. Most NMU of tapentadol occurred via oral routes of administration. Similar to other studies, although it was liked by some, tapentadol did not receive a robust pattern of endorsement for NMU.