OUP user menu

Source of Drugs for Prescription Opioid Analgesic Abusers: A Role for the Internet?

Theodore J. Cicero PhD, Christopher Neil Shores MBA, Alethea G. Paradis BA, Matthew S. Ellis BA
DOI: http://dx.doi.org/10.1111/j.1526-4637.2007.00323.x 718-723 First published online: 1 September 2008


Objective. There has been a sharp increase in the abuse of prescription opioid analgesics in the United States in the past decade. It has been asserted, particularly by several governmental and regulatory agencies, that the Internet has become a significant source of these drugs which may account to a great extent for the surge in abuse. We have studied whether this is correct.

Design. We asked 1,116 prescription drug abusers admitted for treatment, through standardized questionnaires, where they obtained their drugs. We also attempted to purchase scheduled II and III drugs from a random sample of Internet sites offering such sales.

Results. Dealers, friends or relatives, and doctors' prescriptions were listed as a source of drugs with equal frequency (≈50–65%), with theft and forgery far behind at 20%. The Internet was mentioned by fewer than 6% of the total responders. Because these data suggest either lack of availability or that our sample has not yet realized that the Internet is a potential source, we attempted to purchase scheduled II and III opioids and the unscheduled opioid, tramadol, from a random sample of 10% of the sites listing such sales. We were unsuccessful in purchasing a single scheduled opioid analgesic, but found that tramadol, as an unscheduled drug, was freely available.

Conclusions. The assertion that the Internet has become a dangerous new avenue for the diversion of scheduled prescription opioid analgesics appears to be based on no empirical evidence and is largely incorrect.

  • Prescription Drug Abuse
  • Internet Use
  • Opioid Analgesic Abuse


In the past decade, there has been a surge in the abuse of prescription opioid analgesics [1,2]. Dealers, theft, forged prescriptions, and “doctor shopping” are assumed to be the normal source of these drugs. However, there has been considerable speculation, and often definitive declarations, by the Drug Enforcement Agency (DEA), Food and Drug Administration (FDA), National Institute on Drug Abuse (NIDA), the Substance Abuse and Mental Health Services Administration (SAMHSA), and the National Center on Addiction and Substance Abuse (CASA), that the Internet has become a pervasive and stealthy portal of new drugs into the drug-abuse community (see their Websites) [3–6]. Despite these very wide-spread declarations by authoritative sources dedicated in part to informing the public about substance abuse, we are not aware of any evidence that controlled substances can actually be purchased on hundreds of Websites which offer these products. Furthermore, there is no evidence that prescription opioid analgesic abusers actually use this electronic medium to acquire their drugs even if these drugs are freely available.

In order to address these important points, we, first, asked 1,116 prescription opioid analgesic abusers, through standardized questionnaires, where they obtained their drugs; and second, in a semi qualitative study, we attempted to purchase both scheduled and nonscheduled analgesics from a random sample of 10% of the 467 Internet sites advertising such sales during August–September 2006.


Recruitment of Subjects

Subjects were recruited from 85 treatment centers that were private, self-pay, or accepted health insurance to cover their costs and had very limited public support (<20% public funding). We asked more than 600 such centers representing a very broad geographic base nationwide to participate [2], but only 85 were willing to expend the considerable effort required to recruit subjects. Nonetheless, there was a good distribution of urban, suburban, and rural areas across the United States [7]. Each of the treatment specialists was paid $750.00 to recruit their next 50 consecutive patients/clients with a diagnosis of prescription opioid analgesic abuse, to complete a questionnaire which verified abuse or dependence using Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria [8]. The questionnaire covered: demographics; the source of drugs; and drug usage patterns for hydrocodone, extended- and immediate-release oxycodone, morphine, hydromorphone, methadone, fentanyl, buprenorphine, and tramadol products. In terms of the source of drugs, clients were asked to select all of the sources they used in the categories: (1) dealers; (2) friends or relatives; (3) doctor prescriptions; (4) theft; (5) forgery of prescriptions; (6) Internet; or (7) other. All of the completed questionnaires were identified by a unique case number and were sent directly to Washington University School of Medicine. The treatment specialist did not see the detailed responses of their patients/clients, and we received de-identified information. When completed questionnaires were received and processed, the treatment specialists were notified, and they in turn distributed $25.00 gift certificates provided by a national or local retail chain (e.g., Wal-Mart) as compensation to the individuals who had submitted the completed questionnaires.

Internet Purchase of Opioid Analgesics

We also attempted to purchase the most commonly advertised scheduled drugs, OxyContin® (Purdue Pharma, Stamford, CT), oxycodone, Vicodin and percocet, and the nonscheduled tramadol from a random sample of 10% (47) of the 457 Websites that we found were advertising such sales in a 60-day period in August and September 2006. We furnished credit card information, a mailing address, completed online “medical exams,” and when necessary, joined several required “clubs” for the privilege of ordering these drugs. We notified the DEA, FDA, postal service, Federal Bureau of Investigation, and local authorities of our plans.

Patient/Subject Confidentiality

The protocol was approved by the Washington University Institutional Review Board.



As shown in Table 1, our sample was primarily white (81%), the majority was male (57.6%), and the subjects were very well educated: 39% of the male participants and more than 50% of the female participants had a bachelor's or advanced degree.

View this table:
Table 1

Demographics of prescription opioid analgesic abusers

Male (N = 624)Female (N = 459)
Age in years at treatment, mean ± SD33.72 ± 0.4635.19 ± 0.54
Ethnic background (%)
  African American 8.55 9.83
  Other 8.3917.25
Education (%)
  None 7.78 9.47
  High school or GED53.3338.16
  College or advanced degree38.8952.37
  • GED = general education development.

Overall Source of Drugs

Figure 1 shows the proportion of the male and female participants who used the sources identified on the axis. (Note: clients were requested to check all that applied, so the totals are more than 100%.) As can be seen, dealers, doctors' prescriptions, and friends and relatives were listed as the major sources of the drugs, with 50–65% indicating they had used one or more of these sources. Female participants used doctors' prescriptions to a somewhat greater extent than their male counterparts, whereas the opposite was observed for the use of dealers. Next, by a substantial margin, was theft or forgery of prescriptions. The Internet was listed as a source of drugs by a very small percentage of the total number of subjects, and there were no gender influences in this regard. Only 6% of the responders indicated that they used the Internet as one source of their drugs (41 out of 685 subjects: 28 male, 12 female, and 1 not specified). Of the 41 individuals who claimed to have purchased drugs on the Internet, 13 stated that they bought hydrocodone products, 11 OxyContin®, 2 hydromorphone, and 1 each of fentanyl (patch), Ultram, tramadol, and methadone.

Figure 1

The percent of subjects who indicated they used the source of drugs specified in the legend. Subjects were allowed to select more than one source.

Internet Purchase of Opioid Analgesics

We spent in excess of $2,000.00 to order drugs from 47 randomly selected Websites, but did not receive a single shipment of any of the scheduled II or III drugs we requested: OxyContin®, Vicodin, percocet, Tylenol with codeine, and several other drugs, including “European vikodin” (acetaminophen and a narcotic—type not specified). In many cases, we had to first join clubs or members' groups at nonrefundable costs from $179.00 up to $499.00 before ordering, only to find no drugs available after we paid the necessary fee. In addition to outright fraud, “bait and switch” tactics were frequently used. For example, many sites stated that oxycodone, OxyContin®, or Vicodin could be obtained, but upon reaching the point of ordering, we were told that these drugs were currently out of stock, but they could provide the unscheduled Ultram or its generic form, tramadol. Although 6% of the subjects did report that they used the Internet to obtain drugs, only a handful provided a Website for these purchases. We attempted to purchase drugs from these sites and either could not find any such site or no drugs were available. As mentioned above, some sites offered tramadol for sale. We successfully purchased, as advertised, 30 tablets (50 mg) of the prescribed, nonscheduled tramadol from the first site we contacted for $97.59, which is three times its retail or street price [9]. In order to do so, we had to complete an online “medical examination,” the content of which and our responses are shown in Table 2. Apparently, the attending physician believed that the first author required pain relief and the shipment of tramadol was delivered the next day by Federal Express. The package had to be signed for by someone, but not the “patient” requesting the prescription. The manufacturer (name with held at their request) verified that what we obtained was authentic tramadol. We received daily notices by E-mail to reorder the tramadol and we did so once, within 4 days of the original order, and were successful. In this case, we simply had to indicate that we were a repeat customer. We also received weekly phone calls, with the caller ID blocked, and E-mails every other day, urging us to reorder tramadol, for more than 4 months.

View this table:
Table 2

Summary of Internet order form

Personal Details:
Billing Address:
Shipping Address:
Extra Information:Date of Birth
Your Height
Your Weight
Your SexMALE
Is your Personal Healthcare Practitioner aware that you are requesting this medication?YES
Have you been prescribed this medication before?YES
Have you had a physical exam in the last 12 months?YES
Please state the medical condition requiring you to use this medication.LOWER BACK PAIN
Do you suffer from any seasonal allergies?YES
List in detail any allergies you have to medicines.NONE
Are you currently under treatment for any health problems?NO
Are you suffering from high blood pressure?NO
Are you currently taking any prescription or non-prescription medicines?NO
List any significant family medical history.NONE
Are you currently pregnant or have you been nursing within the past 12 months?NO
Will you be taking other medications while taking this medicine?NO
Please let us know whether you are currently taking any opioid medications such as Tylenol ♯3, Percocet, Vicodin, OxyContin® or suffering from a seizure disorder or convulsions.NA
If you have a history of narcotic or opioid use, please indicate the last time you took a narcotic or opioid.NA
Payment Information:


The fact that the Internet is not a significant source of drugs for prescription opioid analgesic abusers is surprising. Our sample would seem to consist of ideal customers: they are well educated, often employed, and presumably have ready access to credit cards and computers. This suggests that the Internet has not been adopted as a source of drugs either because users are unaware of its potential or because there is simply no access to drugs. Our failed efforts to purchase opioid analgesics, although somewhat lacking in scientific rigor, suggest that the latter interpretation—lack of availability—is correct. As such, our findings conflict sharply with claims made mainly by federal agencies, such as the DEA, FDA, SAMHSA, and NIDA. These agencies and other federally funded research and information services, such as CASA, should be providing the most definitive sources for information on substance abuse, but their claims that the Internet has become a dangerous, and significant avenue for the diversion of prescription drugs (see Websites [5,6]) seem to be largely incorrect. These governmental agencies should either provide data to support their assertions or remove unsubstantiated claims from their Websites and written materials.

We can suggest one very large obstacle to the use of the Internet to purchase prescription opioids, even if there were a number of active sites where such purchases could be made. A credit card was required on all sites to purchase the drugs, and the billing address and shipping address had to be specified. Many individuals, including young adults who might still be living at home, might be wary of providing so much personal information to unknown sources on the Internet for illegal trafficking in drugs and be so careless as to have them delivered to their homes.

Whether our purchase of tramadol was legal is unknown. However, we had no prescription, and the “physical exam” we took to gain a diagnosis of pain requiring relief from opioid analgesics was a sham and, certainly, far beneath commonly accepted medical practices. Moreover, the ability to reorder so quickly with no questions asked is sobering as tramadol has the potential for abuse, although it is low [2,9]. Interestingly, however, it appears that the market for tramadol is not abusers since only one person out of 1,116 abusers indicated they used the Internet for tramadol to get high. Nevertheless, if substances with more abuse potential were available, the lack of controls on reorders would represent very ready access to large supplies of drugs. Thus, it is appropriate to worry about the potential use of the Internet as a source of drugs to the illicit market, but our results suggest that this has not yet happened.

The frequency with which doctors' prescriptions served as a source of drugs, at more than 50% of all users, is surprising and unexpected. While some of these physicians could represent unethical and corrupt “script doctors,” it is doubtful that there would or could be dozens or hundreds of script doctors in the regions we examined. Rather, it appears that there is a great deal of doctor shopping or other scams by patients. Physicians, by their own admission, are poorly equipped to recognize substance abuse and have received very little training in this regard [10–13]. Hence, it is difficult for them to discern when a patient is legitimate, may be a substance abuser, or when a legitimate patient begins to abuse his or her prescribed medications. As a result, efforts need to be made to provide drug-abuse training, starting in medical schools and residencies and as part of continuing medical education. There is a difficult balance to achieve, however, in ensuring that all pain patients receive adequate analgesia, while at the same time denying access to often very clever and deceptive prescription opioid analgesic abusers.

There are several limitations in the approach we used in this study. Most prominently, our convenience sample reflects only those individuals who sought treatment for their abuse problems either because of self-recognition or because of referral by family or the legal system. Furthermore, most had the wherewithal to pay for their care either with their own funds or with some form of health insurance, which would favor the demographics we found. This is certainly a limitation with respect to generalizability of our own results to the entire body of prescription opioid analgesic users. However, our sample would seem to be biased toward drug users more inclined to use the Internet, because of their high educational status, relative affluence, and hence presumed access to computers and credit cards. There are other limitations, the most prominent of which is that we used a mailed questionnaire and, hence, our results are subject to all of the limitations associated with this technique, not the least of which is that follow-up questions or clarifications are not feasible [14,15]. Moreover, such surveys for drug-related problems may bias the results toward under-reporting [16,17]. Nonetheless, and given these caveats, our results do not support assertions that the Internet is now a significant resource for obtaining scheduled opioid analgesics.


Despite the assertions that Internet has become a major source of highly abusable opioid analgesics, we found little evidence to support this conclusion in a large sample of patients in treatment for opioid abuse. Furthermore, our efforts to purchase scheduled II and III opioids from a random sample of Websites which offer such sales were unsuccessful, suggesting that drugs are simply not available on the Internet. The nonscheduled opioid analgesic, tramadol, was, however, freely available without a prescription. This raises serious concerns about the medical legitimacy and appropriateness of these Internet sites for both analgesics and other commonly sought and readily available prescription drugs (e.g., Viagra).


Supported in part by USPHS grant ♯DA019908. Theodore J. Cicero is a paid consultant for several pharmaceutical firms.


View Abstract