John Tierney, in a New York Times article entitled “Trafficker or Healer?” recently brought attention to the difficult issues raised by the prosecution of Dr. William Hurwitz. Hurwitz was accused of violating provisions of the federal Controlled Substances Act by “inappropriately” prescribing large doses of pain medications to patients complaining of pain. Cases like that of Hurwitz and dozens of other physicians charged with criminal violations related to their prescribing of opioid analgesics have sent chills through the pain treatment community, and arguably are making it difficult for chronic pain patients to find physicians willing to treat them.
Undoubtedly, the charges brought against many of these physicians are legitimate, but some constitute overreaching on the part of drug enforcement agents and prosecutors and call into question where we draw the line between criminal action and civil negligence in the prescribing of controlled substances. The basis for prosecution in these cases is the Controlled Substances Act, the regulations implementing it, and comparable state laws. In passing the CSA, Congress sought to control illegal distribution of controlled substances without interfering with legitimate medical practices. More recently, the concept of “balance” has been used to provide a common ground for those concerned with diversion and abuse of prescription drugs for pain control and those concerned with improving the care of patients in pain. The term refers to a belief that government policies and law enforcement action to prevent misuse of controlled substances should not interfere with their essential uses for the relief of pain.
The CSA provides that “[e]xcept as authorized . . . it shall be “unlawful for any person knowingly or intentionally . . . to distribute or dispense . . . a controlled substance.” A physician is authorized to prescribe a controlled substance if he is registered by the U.S. Attorney General and complies with the requirements of registration set forth in the implementing regulations. Those regulations specify that “[a] prescription for a controlled substance . . . must be issued for a legitimate medical purpose by an individual practitioner acting in the usual course of his professional practice.” At issue in the Hurwitz case, and others like it, is what constitutes “legitimate medical practice,” which is not defined in the law or regulations. While this definitional issue is often at the heart of the relevant court cases, the more significant question may be whether the current standard for violation of the law – that is, failure to prescribe within “the usual course of . . . professional practice” and for a “legitimate medical purpose” – is appropriate. Arguably, the standard draws the line too far on the side of prosecutions and inappropriately calibrates the balance necessary to achieve the dual goals of prevention of drug diversion and treatment of pain. The case of Dr. Hurwitz illustrates the difficulties of applying the current standard.
In late August 2006, the U.S. Court of Appeals for the Fourth Circuit overturned the conviction of Dr. Hurwitz, a prominent D.C. area pain physician, for prescribing narcotic pain medicine in violation of the CSA. The appellate court vacated the district court’s decision on the grounds that the lower court did not allow the jurors to consider whether Hurwitz prescribed the drugs in “good faith.” Hurwitz, a controversial physician due to his unconventional prescribing, was beloved by many of his patients for treating their debilitating pain when other physicians could or would not. Many of these patients were on a protocol that included very high doses of opioids. Hurwitz was arrested after federal authorities discovered that a number of his patients were attempting to sell drugs that he prescribed. Among other things, he was charged with 58 counts of drug trafficking, “including two counts each of drug-trafficking resulting in serious bodily injury and drug-trafficking resulting in death.” (The quotations about the case, here and below, are from United States v. Hurwitz [4th Cir., 2006].)
The government argued both that the volume of opioids prescribed by Hurwitz vastly exceeded that considered to be within the bounds of legitimate medical practice and that he knowingly wrote opioid prescriptions for addicts and patients he knew were selling the drugs on the street. The defense, in contrast, argued that the dosages prescribed were reasonable for patients with intractable pain and that there was “no medical reason to stop treating a patient for pain simply because that patient may be abusing illicit drugs and that, in some cases, stopping such treatment may even be more problematic.”
A jury found Hurwitz guilty of multiple counts of drug trafficking and he was sentenced to 25 years in prison. Hurwitz appealed the convictions on three grounds, one of which was that the jury was instructed that it could not consider his “good faith” in his prescribing. Hurwitz argued that “his good faith in issuing the challenged prescriptions was relevant to his intent when treating his patients and thus relevant to the jury’s determination of whether he acted outside the bounds of accepted medical practice or without a legitimate medical purpose.”
The Court of Appeals agreed with Hurwitz. Without the good faith instruction, jurors were unable to consider the fact that Hurwitz believed he was treating his patients’ pain. The Court muddied the water a bit, however, when it added a qualifier to its decision – that “good faith” be assessed objectively, not subjectively.The subjective standard looks at whether the doctor actually believed he was acting in good faith – whether he honestly intended to treat pain. The objective standard asks whether the doctor’s belief that his actions were within the bounds of legitimate medical practice was reasonable. (Other federal courts that have addressed the issue have also applied the objective standard. See United States v. Feingold [9th Cir., 2006] and United States v. Williams [11th Cir., 2006].)
In application, the two standards would seem to overlap significantly. In criminal law cases, since jurors cannot get inside the head of the defendant, it is common practice to permit jurors to infer what a defendant intended based on objective data, almost always as explained by the testimony of one or more medical experts. Such objective evidence generally includes whether the physician did a thorough exam and medical history, whether he asked about other drug use or noticed certain “red flags” that would have alerted him to potential abuse or diversion. To the extent that there is a difference between the two standards, it demonstrates the doctor’s dilemma. He is expected to trust his patient, but if he’s found to have been overly trusting because he failed to do “enough” initial or follow-up evaluation and missed the red flags, under the objective standard he risks criminal prosecution and incarceration even if he proves he had the honest intent to treat the patient’s pain.
Doctors who treat pain see a range of patients. On one end of the spectrum are patients who are truly in pain and may need and use significant volumes of pain medication. They do not abuse or divert their medications. Prescribing for this group would not be criminal. On the other end are drug dealers who don’t even pretend to be patients. They may strike a deal with the doctor – drugs for sex or money. Prescribing under these circumstances clearly would be criminal. Bu, also along the continuum are individuals who come to the physician under false pretenses – individuals who pretend to be in pain but actually have no pain and plan to sell the drugs on the street. Some may be good actors. How is the well-meaning doctor to ferret them out? There is no objective test for pain. Determining whether someone is lying is a judgment call often based on seeing the individual and observing his or her behavior. Trying to second-guess a physician confronting such a patient seems unfair. In the Times article, Tierney refers to a recent study by Drs. Jung and Reidenberg indicating how difficult it is for physicians to spot deceptive patients, and he points out that “Doctors are especially gullible because they have a truth bias: they are trained to treat patients by trusting what they say.”
Another complexity is that some patients may be diverting or abusing but also legitimately be in pain. Is it fair to prosecute a physician who prescribes opioids to such patients even if he knows there is a possibility that the patient may abuse or divert the drugs if he is attempting to relieve the patient’s pain? Given that possibility, in the absence of some kickback or tangible benefit (cash, sexual favors, continuing to receive payment for maintaining an unnecessary doctor-patient relationship), or incontrovertible evidence that the doctor has simply exercised no medical judgment at all, it is difficult to justify criminal prosecution of a doctor for his prescribing or dispensing.
While pain treatment advocates may see the Fourth Circuit’s decision in the Hurwitz case as a victory, it’s not at all clear that it has rectified the imbalance of the current law, which seems uncomfortably close to a civil negligence standard. An objective good faith test to determine whether a physician is prescribing within the “usual course of professional practice” and for a “legitimate medical purpose” appears virtually identical to the civil negligence test of whether the physician was prescribing consistent with the current standard of care. In applying the latter standard, inquiries would include whether the physician did a thorough exam, took the patient’s medical history, asked about other drug use, or noticed certain “red flags” that would have alerted a competent physician acting in the same circumstances to potential abuse or diversion.
So what is an appropriate alternative standard? One way to make the distinction clearer between what is required for criminal prosecution and civil liability would be to require that the prosecution establish that the physician knowingly or intentionally prescribed a controlled substance for a nonmedical purpose or a purpose not authorized by law. Under the current standard, as a practical matter, the physician is forced to establish that he or she was prescribing for a “legitimate medical purpose” and was acting in the “usual course of his professional practice.” In order to do this, the defense must bring in medical experts willing to testify that the defendant has conformed to current standards of practice. The alternative standard would force the prosecution to bring in medical experts who would testify that the physician knowingly or intentionally was prescribing for a nonmedical purpose or without a legally valid medical purpose (not whether he was prescribing consistent with a “legitimate medical purpose”). This would require a showing that there was simply no rational medical basis for his prescribing. Evidence of some tangible benefit in exchange for the prescription would also be probative.
Arguably, the proposed standard would help to differentiate the criminal and civil bases for action. This somewhat “higher” standard for criminal prosecution also seems appropriate in this context when one considers the goals of the criminal justice system. Those goals include specific and general deterrence, incapacitation, and retribution. Clearly, we do not want to deter physicians from prescribing opioids for patients who need them. Criminally prosecuting pain physicians for prescribing large doses of opioids in a manner inconsistent with legitimate medical practice will (and has) had an overly broad deterrent effect, reducing the number of physicians willing to treat chronic pain patients. Moreover, incarceration (incapacitation) is arguably not necessary to keep these physicians from doing harm, as the Attorney General can revoke their registration to prescribe controlled substances and the state medical board can revoke or suspend their license. Lastly, retribution would not seem appropriate when there is some evidence of prescribing for a legally authorized purpose and when there are other ways than criminal prosecution by which to deal with physicians who are dangerous or incompetent in their prescribing. In addition to registration or licensure revocation, physicians who prescribe in a manner that is below a professional “standard of care,” and that harms a patient or third party to whom the physician owes a duty, should be liable for civil negligence. These interventions seem better suited to this area of prescribing where physicians are required to make tough judgment calls.
William Marcus, a legal expert in pain treatment issues, and professors Richard Boldt, Andrew Levy, and Sandra Johnson provided comments and suggestions that helped shape this essay.