Home > Encyclopedia of Drugs, Alcohol, and Addictive Behavior > Iatrogenic Addiction

THE RISK OF ADDICTION OR ABUSE

Like any other potential adverse outcome of drug therapy, the prevalence of iatrogenic addiction (drug addiction or abuse during medical treatment) must be determined so that the risk can be assessed by both the practitioner and the patient. An accurate understanding of prevalence, in turn, requires the application of clinically relevant definitions of these phenomena. Unfortunately, there has been little effort to define the addiction syndrome as it occurs in patients, and there is abundant evidence that clinicians commonly use definitions that are inappropriate.

Definition of Addiction in Medical Patients.

Accepted definitions of addiction and abuse (Jaffe, 1985; Rinaldi et al., 1988) have been derived from experience with addict populations. These definitions emphasize that addiction is a psychological and behavioral syndrome characterized by psychological dependence on the drug and aberrant drug-related behaviors. There is loss of control over drug use and evidence of compulsive use. Use of the drug continues, and often escalates, despite overt harm to the user or others. The definitions for abuse project a similar sense and stress the persistence of harmful drug use (Rinaldi et al., 1988) or its deviation from accepted societal or cultural norms (Jaffe, 1985).

The validity of these definitions has not been evaluated in medical populations. Although specific behaviors must be used to establish the diagnoses of addiction or abuse, there have been no studies that assess the predictive value of those behaviors that commonly raise concern in clinicians (Table 1). Some behaviors, such as dose escalation, that strongly support a diagnosis of addiction in an individual who does not have an appropriate medical condition or obtains the drug from nonmedical sources may be more difficult to interpret in patients who acquire the drug from a physician to manage an appropriate problem. Some patients with unrelieved cancer pain, for example, have been said to demonstrate pseudo-addiction—behaviors that suggest addiction but disappear as soon as analgesia (pain relief) improves (Weissman & Haddox, 1989).

In the absence of adequate studies of addiction and abuse in medical patients, the evaluation of drug use in the clinical setting is based on observed

situations. Although some behaviors may provide compelling evidence (selling prescription drugs), most will require astute and often repeated assessments. Any suggestion of aberrant drug-related behavior should impel a comprehensive assessment by the clinician of all aspects related to the patient's medical disorder and treatment plan (Portenoy & Payne, 1992).

The Problem of Mislabeling.

Clinicians often compound the problem of definition by mislabeling patients as addicts without the evidence to support this diagnosis. Such mislabeling increases the perceived prevalence of iatrogenic addiction and unnecessarily stigmatizes the patient.

The most common type of mislabeling confuses PHYSICAL DEPENDENCE with addiction. Physical dependence is a pharmacologic property characterized by the occurrence of an abstinence syndrome following abrupt dose reduction or administration of an ANTAGONIST. Since physical dependence is not apparent unless an abstinence syndrome occurs, and abstinence can be easily prevented, physical dependence is generally regarded as a minor problem in the clinical setting. Although it has been postulated that abstinence symptoms can become conditioning stimuli that contribute to the genesis of addiction (Wikler, 1980), it is evident that physical dependence alone does not produce addiction or abuse. Opioid addicts, for example, may or may not be physically dependent, and cancer patients, who are almost certainly physically dependent after receiving high opioid doses for prolonged periods, almost never develop the aberrant drug-related behaviors consistent with addiction or abuse (Kanner & Foley, 1981).

Studies of Addiction or Abuse in Medical Patients.

Thus, the risk of iatrogenic addiction or abuse can only be determined if proper definitions are developed and applied to patient populations. Few studies have met these criteria, but those that have are reassuring, indicating a very low risk of these outcomes during medical treatment with drugs of abuse.

Surveys of opioid use are most illustrative. Although older studies of opioid addicts suggested considerable risk of iatrogenic addiction, these data have been replaced by more recent surveys of pain patients. Addiction and abuse are vanishingly rare outcomes of opioid therapy for acute and chronic cancer pain (Kanner & Foley, 1981; Chapman & Hill, 1989). Most experts have concluded that the risk of addiction during opioid treatment for cancer pain is so remote that this outcome should not even be considered in the decision to use these drugs. Similarly, the Boston Collaborative Drug Surveillance Project could document only four cases of addiction among 11,882 patients with no prior history of substance abuse who were administered an opioid during hospitalization (Porter & Jick, 1980); a national survey of burn units could not identify a single case of addiction among 10,000 patients who had no history of substance abuse and received opioids for burn pain. Finally, surveys of selected patients with chronic nonmalignant pain also suggest that aberrant drug-related behavior is distinctly uncommon among those with no such history who are administered opioids on a long-term basis (Portenoy, 1990).

Other drugs have not been evaluated as extensively as the opioids. Recent analyses of BENZODIAZEPINE use, however, conclude similarly that addiction or abuse as defined here is a rare outcome among patients with ANXIETY disorders who are administered these drugs by physicians (Woods et al., 1988; Balter & Uhlenhuth, 1991), although many develop physical dependence.

Together, these data indicate that medical patients with no prior history of substance abuse have a very low risk of iatrogenic addiction or abuse when they are medically administered drugs with a potential for these outcomes. This conclusion is consistent with an understanding of addiction as a disorder related to the use of specific drugs, but not inherent in the pharmacology of any. Addiction is presumably determined by an interaction between the reinforcing qualities of some drugs and a constellation of individual factors, including a genetic propensity, psychosocial aspects, and the specifics of drug availability (Jaffe, 1990, 1992; Chapman & Hill, 1989). The evidence suggests that patients who do not demonstrate a proclivity to addiction or abuse by adulthood are extremely unlikely to develop these outcomes during medical treatment thereafter. Furthermore, it is probable that this small risk could be reduced further by strict adherence to guidelines that set parameters of appropriate patient behavior and follow-up assessments. Such guidelines would also facilitate the identification of those occasional patients who develop any addiction problems.

UNDERTREATMENT

Although the conclusion that iatrogenic addiction and abuse are rare, still this appears to be inconsistent with the attitudes held by many healthcare providers and patients. Fear of addiction is commonplace. Consequently, there is evidence that overconcern about addiction adversely influences prescription practices.

The negative effects on patient care produced by an inaccurate estimate of addiction liability are most clearly documented in pain management—inadequate treatment with opioid drugs results in an unnecessarily high prevalence of unrelieved acute pain, especially cancer pain. Concerns about addiction are among the salient factors that contribute to undertreatment (Portenoy, 1995).

CONCLUSION

The data extant indicate that addiction and abuse are rare outcomes during the therapeutic use of opioids and other drugs in populations with no prior history of substance abuse. The intense concern expressed by clinicians and patients alike and the impact of this concern on prescribing practice appear to be disproportionate to the actual risk. To some extent, this may relate to the difficulties encountered in evaluating addiction and abuse in medical populations, or perhaps more likely to the tendency to mislabel outcomes as addiction that do not fulfill criteria for the diagnosis. Although good clinical practice must recognize the potential for addiction and abuse, optimal therapy depends on an accurate understanding of these phenomena and the limited role they play in clinical practice.

(SEE ALSO: Abuse Liability of Drugs: Testing in Humans; Addiction: Concepts and Definitions; Controlled Substances Act of 1970; Diagnostic and Statistical Manual [DSM]; Disease Concept of Alcoholism and Drug Abuse; Opioids and Opioid Control; ; Prescription Drug Abuse; Vulnerability as Cause of Substance Abuse)

RUSSELL K. PORTENOY